Datasets:

kristmh

/

highest_and_high_vs_rest

like 0

note this bug report is for jira service desk server using jira service desk cloud see the corresponding bug report steps to reproduce install jira service desk create a project through jira service desk disable or uninstall jira service desk expected behavior jira would hideremove any configuration regarding jira service desk and would continue to run normally actual behavior jiras dashboards stop working it is not possible to log in but through loginjsp accessing a project is not possible trying to view the permission tab through the project created throws an error workaround option reinstallreenable jira service desk option if unable to log in access jira through navigate to administration » issues » permission scheme delete • anyone from all permissions in all permission schemes option delete these from the database using codesql borderstyledashed delete from schemepermissions where permtype sdcustomerportalonly code if deleting through the database please make sure to create a backup also make sure to shutdown jira first and restart it afterwards notes this seems to be happening because jira service desk does not remove or hide the service desk customer portal access anyone permission and when jira tries to load it the following is thrown in the screen noformatborderstyledashed javalangnullpointerexception at at at noformat this also appears in the logs noformatborderstyledashed error matheus uncaught exception thrown by rest service javalangnullpointerexception at at at at at at noformat

summary jira performance degrades significantly during full and incremental directory sync cpu spiking to page load delay between min environment jira or later jira configured with ad ldap directory type connected steps to reproduce steps to reproduce configure microsoft ad in jira make sure ad has enough users ie users import few users login to jira as local user and be on directory page configure jmeter http sessions using one of the ad user session should repeat contentiously for minute initiate concurrent sessions from jmeter eg browsing boards issues etc after second of jemeter sessions initiate ldap sync login to jira on another browser and keep navigating pages after minute you will notice slowness and ldap sync is running forever expected results directory sync should happen without having performance degradation actual result jira performance significantly degrades when full and incremental sync is happening verification to verify if the instance is affected by this bug collect thread dumps during slow performance as per generate a thread dump reviewing them the caesium thread will contain the below thread over several thread dumps noformat daemon tid nid waiting on condition javalangthreadstate waiting parking at sunmiscunsafeparknative method parking to wait for a javautilconcurrentlocksreentrantreadwritelockfairsync at at at at at at at at at source at at at at at at at at at at at at at at at at at at at at at at at at at at at at at at at noformat and there will be a number of tomcat worker threads waiting on futures inside that cache as per the attached screenshot essentially the sync is invalidating a cache that a large number of actions rely on and when its invalidated those threads will block until its repopulated workaround if using active directory set the jira system property noformat dcrowduselegacyadincrementalsynctrue noformat this workaround will mean that users who no longer exist in ad but own content in jira will not be deleted from the cache on an incremental sync only thus not triggering this issue a full sync will still be affected however note on partial fix we significantly reduced the performance problem by resolving issue which was about skipping active user counting in cases when it was not required after gathering feedback we decided to reopen this issue to look further into how can we improve user synchronisation performance

when used the redirect macro and export space to pdf files this error appears in pdf file error formatting macro redirect javalangnullpointerexception steps to reproduce go to a page and enter edit mode click the insertedit macro icon and select the redirection macro from the list of macros on the insert redirect macro dialog enter the name of the page in the redirect field then click insert to create a pdf of that space select browse advanced pdf export then click export this is the result using redirect macro if i export directly without the redirect macro this message doesnt appears redirectjpg

issue summary bitbucket pipeline is unable to pull images as service from the nexus registry steps to reproduce pull an image as service from nexus registry in bitbucket pipeline the following error occurs codejava failed to resolve reference not found code expected results the image from the nexus registry should be pulled as service without any issue in the bitbucket pipeline actual results the image can not be pulled as service in the bitbucket pipeline workaround the nodes in our kubernetes build cluster are being run with containerd rather than docker it seems that recent versions of nexus at least and doesnt support containerd please file a support ticket to be excluded from our migration and consider raising a support ticket with nexus to support containerd

summary the epic link field is not populated when create issue in epic from issue preview screen in the jira project view this prevents tickets from being created as the epic link field is required but cannot be entered manually environment jira software cloud steps to reproduce navigate to the issues view for a software project with an epic select the epic from the issues list so that the issues is previewed to the right of the screen click on the to open the create issue in epic pop up dialog fill in the required fields and click create screen shot at ampngthumbnail expected results the epic link field is populated and the issue is created actual results the error colorredepic with key does not existcolor appears and the issue cannot be created screen shot at ampngthumbnail this does not occur when the create issue in epic happens when the issue is opened outside the preview screen so there are a few options search for the epic issue in the screenshot to open the browse issue screen and create the issue in epic from there browse search browse search screen shot at ampngthumbnail click on the issue link on the preview window to open the issue and create the issue in epic from there click link click link screen shot at ampngthumbnail

description jira targz source code not able to download in i issue only happen for jira targz download link zip download link and other version targz are working perfectly results source code able to download results code this xml file does not appear to have any style information associated with it the document tree is shown below nosuchkey the specified key does not exist code

hiwe are currently experiencing an issue with a post service not being called when a user new changeset is pushed on your serviceafter multiple trial i come to the conclusion it only produce the issue while pushing with the ssh protocol not the httpsi know if it is related but i have an error message at the end of the sshpushpushing to sshhgbitbucketorgwearesandboxsearching for changesremote adding changesetsremote adding manifestsremote adding file changesremote added changesets with changes to filesremote bbacl ahsx is allowed accepted payloadremote error changegrouplastupdated hook raised an exception exception got an unexpected keyword argument

when doing certain operations in a repo ex fetch pull etc or sometimes when the automatic refresh takes place the left side menu showing my repos branches and the remote repos branches collapses this is very annoying since i usually change between branches often and having the list of branches collapsed all the time is very annoying this is something that changed in the latest releases the first version did not have that bug and the version never had any issue similar to that

manually created review comment sends notification email to revieweres but the one created through rest does not

it is possible to fake log entries in fisheyecrucible logs by sending specially crafted http requests containing a newline character for example going to the url will cause the following to be logged code info fisheye no such repository asd fake log entry referernull code

you shoudldnt be able to expandcollapse frxs in singlefileview in fullscreen mode those icons dont render when not in fullscreen mode singlefileview they keyboard shortcuts to expandcollapse also have the same behaviour ie work in fullscreen mode when they shouldnt but dont work in nonfullscreen mode

when user is required to confirm the password confluence always checks the entered password against the internally stored userpassword if an instance is configured to use custom authentication which is different from atlassianuser the password validation will fail resolution this is fixed in confluence and later versions we check if the confluence instance is configured to use a nondefault seraph authenticator and automatically disable the functionality that relies on password confirmation web sudo captcha password confirmation on email change to overwrite this behavior use passwordconfirmationdisabled flag if you set this flag to false than even if you have a custom authenticator password confirmation will still work as configured and will try to validate the password against the user managment configured through atlassianuserxml note that web sudo and other password confirmation screens should probably be disabled if you use an sso authenticator confluence is typically not able to verify a users password so we recommend using some other mechanisms for your administrative security

hi atlassian it appears that the problem reported in is back with jira but only for one ressource environement ibm websphere oracle side effect the calendar is in transpency calendarwithoutcsspng root cause the request for its css calendarcss fails with a first analyse the failed request is but the others also fail however the other one succeed it appears that the usage of comibmwswebcontainerinvokefilterscompatibilitytrue in websphere web container has not effect for my issue after some tests remote debug it appears that resourcedownloadutils defined in urlrewriter configuration is with an empty browser cache well called and served for but not for fileserverservlet is also well called and well served for well called and not served for but not called for i tried to call direclty but no return occurs finally when calling directly the css is provided for an nonidentified reason with the following servlet codexml calendarcss stylescalendarcss calendarcss includesjscalendarskinsaquathemejsp fork false code and codexml globalcss stylesglobalcss globalcss stylesglobalcssjsp fork false code the first fail in fowarding to the jsp where the other success i do not find real link with or help

using the jira with internet explorer is know completely broken there are so many layout issues and also lots of functions that dont work it seams to me that jira has lost complete support for ie with jira most of these things worked fine i think this is a mayor problem because you will see that a lot of enterprise customers still use ie my company has arround windows xp ie and is starting now with migration to windows with ie this huge project will take to years

note this bug report is for confluence server using confluence cloud see the corresponding bug report steps to reproduce subscribe to a calendar the subscription section will look like this change language setting to deutsche subscribe to another calendar the subscription section will look like this noticed that there are two subscription sections now one in english the other is in deutsche workaround uninstall team calendars plugin and install team calendars plugin instead

problem definition jira servicedesk mail handler has difficulties processing msg files from outlook the attachment wont appear on tickets test with gmail returns no issue and works as expected steps to reproduce create email in outlook attach msg file send to jira servicedesk mail handler result ticket should be created with msg file attached actual result ticket will be created without msg file

note this bug report is for jira cloud using jira server see the corresponding bug report paneli found a number of problems with jira ondemand after upgrading to chrome on the issue screen if you hover your mouse over the issue title some buttons disappear drop downs on modal windows eg new issue window are a bit strange the dropdown button is taller than the dropdown itself will add as i find more issues

summary feature prefix is not added to the branch name automatically when bugfix branch model is disabled in repository settings in bitbutcket server steps to reproduce jira integrated with bitbucket server go repository settings in bitbucket server select branching model check use custom settings in project settings inheritance remove the check in bugfix in branch prefixes press save go jira open an issue click create branch in development feature is selected in branch type but feature prefix is not added to the branch name expected results expected branch name actual results actual branch name createbranchbitbucketjpgthumbnail reselect feature branch type in branch type dropdown list

note this bug report is for confluence cloud using confluence server see the corresponding bug report panel

we can not access out hosted studio solution httpdigidatajiracom

reproduce step open repository with fullscreen open other repository by doubleclick local repository list actual result sourcetree app crashed expected result in my memory open repositories with tab on fullscreen mode ive attach crash log

we like to use labels starting with a a la twitter hashtags but these do not show up in your labels gadget perhaps theres a good reason for this obviously this is not a biggy

hi we are facing issue while exporting any confluence space to the xml format step to reproduce browse the space whose backup has to be taken go to advance tab click the export space chose xml output and export the space after some time it will gives an error attaching screenshot for the error message error message exception initializing proxy comatlassianconfluencespaces i have checked log file their is noting in it regards

sybase doesnt seem to like you putting null in a clob column

when perform upgrade to greenhopper in jira using oracle database below error has been found noformat error leon could not migrate ranks from old rank field to new rank field info leon undo class noformat and noformat info leon undo done class error leon upgrade failed null javalangreflectinvocationtargetexception at method at at at noformat end up with noformat caused by comatlassianactiveobjectsinternalactiveobjectssqlexception there was a sql exception thrown by the active objects library database nameoracle versionoracle database enterprise edition release production with the partitioning real application clusters olap data mining and real application testing options minor major driver nameoracle jdbc driver javasqlsqlexception functie wordt niet ondersteund at at at method noformat as far as we discovered this only happened in oracle database

i just clicked the download embedded button and the app crashed

as soon as i click the checkbox of a modified file in the unstaged files list on the log history tab source tree crashes the following stack trace appears in the log error fatal error systemnullreferenceexception object reference not set to an instance of an object at sourcetreeviewmodelrepotabviewmodeltogglestagingifilestatusrecord filestatusrecord in cprojectsbitbucketorgatlassiansourcetreewinprodsourcetreeapiuiwpfviewmodelrepotabviewmodelcsline at sourcetreeviewmodelutilscommandsinkexecutecommandicommand command object parameter boolean handled in cprojectsbitbucketorgatlassiansourcetreewinprodsourcetreeapiuiwpfviewmodelutilscommandsinkcsline at s executedroutedeventargs e in cprojectsbitbucketorgatlassiansourcetreewinprodsourcetreeapiuiwpfviewmodelutilscommandsinkbindingcsline at systemwindowsinputcommandbindingonexecutedobject sender executedroutedeventargs e at systemwindowsinputcommandmanagerexecutecommandbindingobject sender executedroutedeventargs e commandbinding commandbinding at systemwindowsinputcommandmanagerfindcommandbindingcommandbindingcollection commandbindings object sender routedeventargs e icommand command boolean execute at systemwindowsinputcommandmanagerfindcommandbindingobject sender routedeventargs e icommand command boolean execute at systemwindowsinputcommandmanageronexecutedobject sender executedroutedeventargs e at systemwindowsroutedeventargsinvokehandlerdelegate handler object target at systemwindowseventrouteinvokehandlersimplobject source routedeventargs args boolean reraised at systemwindowsuielementraiseeventimpldependencyobject sender routedeventargs args at systemwindowsinputroutedcommandexecuteimplobject parameter iinputelement target boolean userinitiated at sourcetreeviewfilelistsabstractcheckboxfilelistviewcheckboxonpreviewmouseleftbuttondownobject sender mousebuttoneventargs e in cprojectsbitbucketorgatlassiansourcetreewinprodsourcetreeapiuiwpfviewfilelistsabstractcheckboxfilelistviewcsline at systemwindowsroutedeventargsinvokehandlerdelegate handler object target at systemwindowseventrouteinvokehandlersimplobject source routedeventargs args boolean reraised at systemwindowsuielementreraiseeventasdependencyobject sender routedeventargs args routedevent newevent at systemwindowsroutedeventargsinvokehandlerdelegate handler object target at systemwindowseventrouteinvokehandlersimplobject source routedeventargs args boolean reraised at systemwindowsuielementraiseeventimpldependencyobject sender routedeventargs args at systemwindowsuielementraisetrustedeventroutedeventargs args at systemwindowsinputinputmanagerprocessstagingarea at systemwindowsinputinputprovidersitereportinputinputreport inputreport at systemwindowsinterophwndmouseinputproviderreportinputintptr hwnd inputmode mode timestamp rawmouseactions actions x y wheel at systemwindowsinterophwndmouseinputproviderfiltermessageintptr hwnd windowmessage msg intptr wparam intptr lparam boolean handled at systemwindowsinterophwndsourceinputfiltermessageintptr hwnd msg intptr wparam intptr lparam boolean handled at hwnd msg intptr wparam intptr lparam boolean handled at o at systemwindowsthreadingexceptionwrapperinternalrealcalldelegate callback object args numargs at msinternalthreadingexceptionfilterhelpertrycatchwhenobject source delegate method object args numargs delegate catchhandler

panel bgcolorffffce the disabled modules that you see in will not affect the running of jira service desk those modules are there for upcoming features seeing those disabled modules is normal panel while having jira service desk with version below installed update the plugin and some modules will be disabled and could not be able to be enabled so far by having these modules disabled it is not hindering with servicedesk operation and reenabled jira service desk would not enabling these modules as well stack trace when upgrading jira service desk where some modules are actually being skipped noformat jira servicedesk devversion built stopped info admin removed plugin comatlassianservicedesk info admin waiting for enabling plugins info admin plugins that have yet to be enabled seconds remaining threadpoolasynctaskexecutorthread error admin unable to load class comatlassianservicedeskpluginsautomationapiconfigurationrulesetinputbuilderservice for component importation purposes skipping threadpoolasynctaskexecutorthread error admin unable to load class comatlassianservicedeskpluginsautomationapiexecutioncommandruleexecutioncommandbuilderservice for component importation purposes skipping threadpoolasynctaskexecutorthread error admin unable to load class comatlassianservicedeskpluginsautomationapiexecutionmessagerulemessagebuilderservice for component importation purposes skipping threadpoolasynctaskexecutorthread error admin unable to load class comatlassianservicedeskpluginsautomationapiexecutionwhenhandlerwhenhandlerprojectcontextservice for component importation purposes skipping threadpoolasynctaskexecutorthread error admin unable to load class comatlassianservicedeskpluginsautomationapiexecutionwhenhandlerwhenhandlerrunincontextservice for component importation purposes skipping info admin plugin comatlassianjirajiraemailprocessorplugin is now enabled info admin plugins that have yet to be enabled seconds remaining info admin plugins that have yet to be enabled seconds remaining info admin plugins that have yet to be enabled seconds remaining info admin plugins that have yet to be enabled seconds remaining threadpoolasynctaskexecutorthread info admin jira servicedesk devversion built plugin starting noformat the screenshot of some modules is not completely being enabled in jira manage addons page modulpngthumbnail

note this bug report is for jira service desk cloud using jira service desk server see the corresponding bug report panelim currently using if i click the edit form modify the description and click the check button it doesnt give an error but the new value is not saved i tried both chrome and firefox editing it from the main screen works fine though version

not able to push to bitbucket with ssh keysi have added ssh key to my accountgit push u origin all gives in following errorcodebashthe authenticity of host be establishedrsa key fingerprint is you sure you want to continue connecting yesno nocodesshadd l gives homekrypdonsshkrypdon rsacode

update this to take in the second commit

i created an application link between my confluence and jira instance and included jiras activity stream gadget as an external gadget when i add the gadget on a confluence page the jira data is correctly shown however the point to the following url instead of this makes it impossible for users to directly navigate to the correct jira ticket i seems that confluence takes its own base url to create the full jira ticket url instead of the one from jira as a test i tried to change the base url from confluence and indeed this also affected the jira issues links maybe this issue is fixed in a later version but could you provide me with a workaround for now kind regards benji

steps to reproduce share an issue with yourself using your account name no email share the same issue with yourself using your email receive notification configure profile to notify yourself on your own changes share an issue with yourself receive notification explicitly sharing an issue with yourself should not be controlled by the notify me on my changes setting

when i installed the beta none of the bookmarks from the release were migrated to the beta i had folders defined and several repositories bookmarks under each one i had to reenter them all in the beta

dear bitbucket supporttoday i was working at one of eformations project and did a commit as usual but in some mysterious manner the commit user is displayed als cossig instead of me dominik piekarskican you give me some advise how this is possible i suppose this is a pretty nasty bugregardsdominik piekarski

perhaps nothing should highlight at all

when i select that i want to view the log for the current branch i still see the other branches the only difference between current branch and all branches is that the branch tree changes configuration switching between these settings does not help

the order of child pages show in the api does not match the order show in the confluence steps to reproducecreate a page and under it create a series of child pages with names starting with different characters you will see that in the confluence ui they are ordered alphabeticallynext load the child pages api for that parent page if the parent page you created has the id then the url would be expected behaviourthe order of pages matches the behaviour experiencedyou will see that the order of the pages does not match the ui and that the extensions position property is none

paneltitlenot a jira bug borderstyledashed i this is not a jira core bug ticket created to track the problem and for transparency the problem is a combination of different factors java platform and libraries limitations that currently exist panel issue summary under a very high load and a large number of lucene operations on a large lucene index jvm starts having contention in reading javalangthreadlocal variables to be specific javalangthreadlocalthreadlocalmapgetentry and javalangthreadlocalthreadlocalmapgetentryaftermiss calls become very contended which leads to high cpu usage and high la in short it’s a combination of factors high allocation rate of threadlocal caused by lucene lib its tracking index segments in orgapacheluceneutilcloseablethreadlocal and as consequence very high number of javalangthreadlocalthreadlocalmapentry in the corresponding javalangthreadlocalthreadlocalmap slow gc collection of threadlocal since they are weakly referenced a feature of modern gc see high pressure on threadlocal by doing high concurency reads from cache jira uses ehcache which uses reentrantreadwritelock internally which uses threadlocal steps to reproduce large jira instance with large lucene to have many segments apply high load a lot of lucene operations large number of cache reads eg cachingprojectroleandactorstoregetprojectrole expected results application is stable under load actual results application is not stable large cpu usage growing number of javalangthreadlocal and corresponding javalangthreadlocalthreadlocalmapentry notes about the symptoms example of the stacktrace collected during the peak time noformat daemon tid nid runnable javalangthreadstate runnable at at at at at at at at at at at at at at noformat count of stack in the threaddumps noformat grep c javalangthreadlocalthreadlocalmapgetentryaftermiss jirathreads noformat memory allocation class number of objects size javalangthreadlocalthreadlocalmapentry javalangthreadlocalthreadlocalmapentry javalangthreadlocal other notes reference to similar problems with elastic search opendj bq when shortlived threadlocals are created and die at a rate linear to application throughout this leads to threadlocal maps inactive threads holding thousands or millions of entries to extremely long collision chains and to situations where half the cpu is spent walking those chains in threadlocalget note on threadlocal threadlocal variables use weakreferences as internal implementation which are subject to different cleanup rules than normal references on we have a time budget for maximum gc pause a soft goal and jvm may opt to delay the collection of weakreferences to a cleanup of tenured space to meet that goal it will only clean up a subset of cleanable weak references in young gc this may cause a buildup of entries in the threadlocalmaps of threads the following jvm parameter will give more details in gc logs about how much weakreferences are cleaned up xxprintreferencegc workaround a couple of options are available here upgrading to jira due to changes in lucene and jira it produces less threadlocal tuning give it less aggressive goal to allow more weakreferences to be collected switching to cms based on elastic search examples resolve the problem as it collects weakeferences more aggressively its an old gc algorithm may have more frequent peaks and longer average collection time

note this bug report is for confluence server using confluence cloud see the corresponding bug report panelive noticed some answers go missing off my stats and discovered that the questions themselves have been deleted is this desired behavior for confluence questions is this desired behavior for answersatlassiancom

when user copy a sentences and paste it on another sentences with a different font style the font style will be retain even after the word is overwritten this only occur when you start highlighting directly from the end of the sentences that you wish to paste or overwrite then paste it with the word you wish to copy over please refer to the screen record for the step to properly replicate the bug

to reproduce pick a studio instance with multiple projects create an issue in one of them say issue commit a change set that spans multiple projects either using a client that supports xproject commits like subclipse or just by checking out the repo from nb you may need to change permissions to get commit rights to and with a message like resolve time comment this will appear twicethe issue will get two comments of time logged and you will receive a failure notification from the second attempt to resolve the issue

when i am trying to access my repository via an connection the server keeps telling mecodecommandlinewarning permanently added the rsa host key for ip address to the list of known hostsfatal could not read from remote repositoryplease make sure you have the correct access rightsand the repository existscodebut ok if i only use the to do so

note this bug report is for confluence cloud using confluence server see the corresponding bug report panelsteps to reproduce on the latest version of open an inline close highlight text to create new inline comment expected the sidebar is opened and the cursor is in the inline comment editor actual result the sidebar appears then quickly open another inline comment then click on reply expected the editor appears and the cursor is inside the editor actual result the sidebar disappears

im using atlassian connector for intellij idea and trying to create review on code calls api method post http header values applicationxml i selected parameters project repository not related to selected change set change set from different repo than selected and got error comatlassiantheplugincommonsremoteapiremoteapiexception server returned http internal server error

if you choose an event jql but do not select any events and then only tie the webhook to workflow post functions jira will follow the jql chosen and will only fire the webhook for issues that are within the range of search but the jql is not shown in the webhooks page until you click on edit this makes harder to view what is the jql in use for that webhook steps to reproduce create a new webhook add a jql in the issue related events and do not select any events tie the webhook to some workflow post functions save the webhook expected results jira displays the jql actual results you are only able to see the jql if you edit the webhook

issue summary noting in our documentation search syntax for text fields we state quote note all query terms in jira are not case sensitive quote however when using the order by function on a text field the sorting appears to be using a ascii sort function which causes the sorting of uppercase and lowercase letters separately and sort punctuation marks and symbols in various places between numerals and letters for example z precedes a a reference to ascii ordering can be found at using the order of character sets designed for computer sorting does not necessarily produce a sort sequence that is acceptable for human readable fields such as a text field where having values such as in the fields would be sorted as in ascii so that the upper case a b are separated from lowercase a b where alphabetically the uppercase and lowercase letters should be grouped together steps to reproduce create multiple issue adding one of these values in each issues summary a a b b z in jql search add in order by summary expected results the issues are sorted with alphanumeric sort order of special characters then numbers then letters a a b b z actual results issues sorted by ascii value priority a b z a b screen shot at pmpngthumbnail in ascii sort the special characters for a dash and underscore occure between uppercase and lowercase and are not grouped with other special characters causing an additional layer of confusion workaround currently there is no known workaround for this behavior a workaround will be added here when available

issue summary pages in legacy editor when mentioning a user while editing the page you are able to strikethrough the username as it inserts the username as a text with a hyperlink to the user profile in preview mode or after publishing it instead of showing username it shows username without the strikethrough environment confluence cloud legacy editor steps to reproduce create a page and mention user username select the username and add the strikethrough option publish or preview the page expected results the page should be the same in edit vs publishedpreview version workaround currently there is no known workaround for this behavior

i enabled the new source browser experience when i go to view the source code i can drill down thru our directories and see the files in a folder with files however whenever i try to access a folder with files the working circle icon just spins it has been spinning for more than minutes i get the same results in both ie and chrome i have also tried it in multiple of our mercurial repositories

summary the application navigator is not working on confluence on certain instances where the application navigator have the home option on the list the application navigator is not loading on confluence and also is throwing an error on the console screen shot at pmpngthumbnail codejava uncaught typeerror cannot read property push of undefined at e at htmlanchorelementt at htmlanchorelementdispatch at htmlanchorelements code this issue is reproducible accessing confluence on the instance and trying to open the application navigator the mouse pointer spins and theres no response the affected version is workaround none

note this bug report is for confluence cloud using confluence server see the corresponding bug report panel

if you run some tasks from comatlassiansalapischedulingpluginscheduler them will be runned with webappclassloader if your plugin job try to check merge vetos for other pull request via comatlassianstashscmpullmergerequestcheckservice in this job thread then all mergechecks will executed not in httpthreadsxxx bu in other threads and then if you try to serialize listhashmultimap in some of this mergechecks other plugins you may get strange exception code warn javalangclassnotfoundexception comacmestashpluginothercontentsomekeyinsetmultimap at at method at at at at at at at at at at at at at at at at at at at at at at at at skipped at at source at source at source at code the bug only in this thread the same check from web ui work fine i think that pluginscheduler load tasks in not valid classloader

note this bug report is for jira cloud using jira server see the corresponding bug report panelreproduction stepsload refresh the view issue page for an issue on latest jira master ondemandwhat happens the following js error is returnedcodetypeerror ngooglemapsload is not a functioncode as a result clicking on value for location custom field doesnt show google mapenvironment to reproduce problem jira ondemand version browser and os environments browser os firefox mac os x safari mac os x chrome mac os x ie windows service pack ie windows professional service pack ie windows version build as google maps is being loaded on jiraeventsnewcontentadded which is now being fired multiple times per page google maps is potentially being loaded twice by jira suite utilities resulting in the js error occurring and the inability to view google maps of location fields a jira developer has listed the following for jira suite utilities to investigate in as potential methods to rectify the issue please do not hesitate to mention or contact him if you have any questions regarding these use a namespaced jsonp callback for loading google maps right now ongmapsload is being used and could conflict with other plugins check the reason parameter when listeing to jiraeventsnewcontentadded on kickass we fire that event a few times per page and probably not all of them should trigger a google maps load check that jira suite utilities is not trying to load google maps again if they have loaded it once but they are still waiting for the network update the code to use googleloader instead loading gmaps directly

copied from this might be by design but when a user adds individuals or groups to the edit space permissions page they are immediately granted view privileges if a user hits the cancel button on the page it only resets the checkboxes it does not revert to the original permissions state from before the operation began while it is easy to remove these individuals and groups after the fact it creates a dangerous expectation when administrators are learning the interface as they might hit the cancel button and immediately navigate away from the page before noticing that the permission changes were added normally i could even consider this part of the learning curve for confluence administrators but with my endusers starting to look into personal spaces it is starting to become a security risk would it be possible to save the permissions configuration when an edit operation is performed and have the cancel button revert to that

on some oses the user created by jiras installer has the shell binfalse this causes startjirash to fail if it cant inherit shell str install jira on an ubuntu box reboot or shut down jira use sudo service jira start to attempt to start jira expected outcome jira starts actual outcome output from the jira init script stops after noformat to run jira in the foreground start the server with startjirash fg server startup logs are located in optatlassianjirabinlogscatalinaout executing using dedicated user jira noformat and jira does not start root cause startjirash contains the following stanza codelanguagebash if then sucmdsbinrunuser else sucmdsu fi if then sucmd m jirauser c prgdircatalinash run else sucmd m jirauser c prgdirstartupsh fi code unfortunately the service wrapper on some oses including ubuntu clears the shell environment variable causing su to look up the shell in etcpasswd where its binfalse startupsh never runs workaround manually start jira each boot using etcinitdjira start not service jira start

note this bug report is for confluence cloud using confluence server see the corresponding bug report paneluploading a pdf attachment using attach file browse button corrupts the file as it is uploadedthe resulting file can not be view from the browser adobe reader reports the file is damaged and could not be repairedthis doesnt happen with all pdfsi downed using a binary ftp transfer from the confluence data folder to my local machine and that file is corrupt so the corruption mustoccure during the uploadsave to diskthe file thats having problems is an strategic corporate doc so i cant include it in this posting

the problem is that when you first hit auth you get a new sessionbut nothing ever puts a user in it then when come around again to go to issue seraph sees that you have a session but finds no user in itthen it doesnt bother to look at the cookie seraph should probably be looking at the cookie in this case rather than the session but we do it in that order because we optimise for the common case jirarestplugins problem is that using loginservice doesnt do all the same stuff that the defaultauthenticator does we need to switch to the da or something similar

issue summary inserting excel files in xlsx format with charts into a confluence page via office excel macro display error viewing macro viewxls in page instead of file preview steps to reproduce attach probfilexlsx to a confluence page insert office excel macro into the page and choose probfilexlsx save the page and note the ui error that shows instead of the file preview codenone error rendering macro viewxls comatlassianpluginsconversionconvertconversionexception javalangclasscastexception javalanginteger cannot be cast to javalangdouble code open probfilexlsx and delete the chart insert office excel macro into the confluence page and choose the new problem file with the chart deleted note the file preview shows as expected expected results the file preview should display in the office excel macro with or without a chart actual results the file preview does not display and an error attached is shown instead workaround no known workaround

the viewattachmentservlet method setresponseheaders does code final string codedname jiraurlcodecencodeattachmentgetfilename true responsesetheadercontentdisposition disposition filename jiraencoding codedname code which results in a header that looks like quote contentdisposition inline quote the problem is that browsers appear to have pretty poor support for the standard on this this judgment is based on some quick preliminary testing by myself that confirmed what is on this page of crossbrowser compliance testing in order of level of support it appears to go opera is nearly perfect firefox misses a few edge cases and then a threeway failtie for google chrome safari and ie they fail in a lot of ways but the big problem is that jira is passing back an encoding character set and only opera and firefox can handle it if we use the noncharset encoding format of quote contentdisposition inline filenamea filename with spaces quote then safari at least i didnt try ie because im lazy appears to do the right thing in terms of customer experience this means something as simple as filenames with spaces in it dont work the way they expect unless they are using firefox and even firefox has bugs where it handles it differently depending on whether you click on the link directly or rightclick and save link as obviously we need more testing i was only checking filenames with spaces for instance and some selenium tests around all of this is needed but if this is true then i dont see how including the encoding is terribly useful to our customers it seems like a better approach might be to do browser sniffing andor check the filename encoding and only send an encoding if its nonascii essentially something like what chris owen suggests in which is a confluence bug report about the same kind of thing

if a git repository hosted on bitbucket contains filespaths that use international characters such as ä and ö the corresponding file names are corrupted eg ö gets replaced with ├╢ in a zip package download of the repository that bitbucket created a public repository in bitbucket that should show you the problem

issue summary when using the canadian french keyboard the doesnt work on confluence environment cloud windows steps to reproduce set your keyboard preference to canadian french edit a confluence page try to insert the on the page ctrl alt expected results after using the keyboard commands the character is inserted actual results confluence formatting is changed notes in other products it is working like jira for example workaround copying the from an external font and paste it in the confluence page

from dashboardcachingmanagerimpl uses planmanagergetallplansclass to fetch list of plans to cache unfortunately that method is filtered and returns only plans visible to anonymous users this is a problem when plans are configured to be not visible for anonymous users as they wont appear on the dashboard even for logged in user with proper privileges

note this bug report is for confluence server using confluence cloud see the corresponding bug report panelfind and replace fails in after you input your text in both boxes the find replace and replace all buttons are still greyed outworks in chrome firefox and confluence ondemandworkaround close the toolbar with the x on the right and then hit ctrlf again

on opening the jira on mozilla the content of page are not displayed completly the options for changing project issue navigator filtering options etc are not displayed it keeps displaying loader sign when user click on topleft logo of project showing error with exception when user click on link agile under my profile page was opertaing systemscreenshot attached for reference

the problem is actually in osworkflow as descibed here

create a user user in a crowd login to bamboo and use the crowd server as the remote user management system in bamboo attempt to create a user with user lower case expected error message to be displayed saying that user already exists actual code javalangreflectundeclaredthrowableexception at source at at at code

it is possible in ondemand to link an jira issue to a certain page in confluence by hitting more actionslink however if the name of the page contains a colon which in the url will be translated to an error occurs workaround change the in the url to the colon character and link them thereafter but the app should be able to handle this without changing the url

application crush when exists especial chars like á in the filename

to be clear the following issue makes sourcetree completely unusable i will be downgrading back to until it is resolved after upgrading to sourcetree i get a consistent problem where the sourcetree file status tab randomly shows local change diffs against some random old version of the file instead of against the most recent version this leads to completely misleading diffs that dont reflect the actual changes that have been made if i bring up an external diff of the file it is also against the wrong version so clearly sourcetree is choosing the incorrect version at a low level somewhere stranger still if i restart sourcetree often it will start showing the diff correctly only to then break again as i browse around on other changed files in the file status ui i had never seen this in any version of soucetree prior to it literally started happening the instant i upgraded from to

there is a graphical whitespace below every jira portlet in confluence when viewing w internet explorer

this is happening now only on jira this can be seen on qasac at the moment

note this bug report is for jira service desk cloud using jira service desk server see the corresponding bug report summaryattempting to configure mail in a shared mailbox setting fails with an authentication steps to reproducethis was reproduced with however will also occur on other mail servers setup a shared mailbox attempt to setup an email channel using jira service desk with the below configuration username password expected resultsjira service desk is able to connect to the actual resultsthe below error is presented in the logs and the authentication error captainplanet unable to connect to the server at due to the following exception unfortunately jira service desk couldnt connect to the mail server here is what the mail server said authenticate the below error can occur during shared mailbox not workingjpgthumbnail

the tasklist plugin doesnt export to pdf properly in confluence the macro used to display the creator completed and priority but now only shows the title of each task installed version on dynamic tasklist latest version stable is is

we have had reports that the search query string is not correctly xml encoded before being redisplayed to the user

exporting a pdf does not render the following correctly it shows up as where username is crossed out with no linkcamelcase elements render as and not the url

if you are already logged in and you ask confluence for a url like confluence will inform you you do not have permission to access displaytesthome to access this page please log in as a user with sufficient permissions even if you do in fact have the appropriate permissions steps to reproduce copy a url to a page log out paste url into browser it should redirect you to loginactionosdestinationpathtopage copy that url login paste url back into browser expected output contents of page real output error regarding permissions this was duplicated in and

if you assign an issue with estimate to an epic and then resolve the issue the epic isnt updated accordingly we currently check whether the issue is assigned to the epic and if so update the epic panel we also have to update the epic panel if the issue isnt on the board anymore

add public bitbucket repo add the postcommit hook to the repository commit and push a changeset with a valid issue key on message navigate to the issue and open the bitbucket tabexpected new commit to appear actual commit does not appearnotes after minutes of refreshing and waiting for the commit i force synced and the commits appeared this may be related to if there is a failure to connect to bitbucket at the moment the postcommit occurred the commits will be effectively lost until a force sync

summary unable to attach files in jira issue attachment thumbnail already stored error environment jira core cloud steps to reproduce open any jira issue click on browse a file and attach any filetype the upload will result in an error attachment thumbnail already stored expected results attachment should be uploaded successfully actual results the attachment upload will result in an error attachment thumbnail already stored notes jira is trying to store the attachment using an existing attachment id from mediastoreattachments workaround contact attlassian support

after deploying jira under weblogic and connected to a mysql database jira runs extremely slow when creating or doing the find issue feature and then spits out this errori have checked and rechecked all my settings and recompiled the app a number of times i cant figure out why this is happening it seems to be completely random and everything on my weblogic server looks okas you can see im using the standard jdk with because the jrockit jdk was no betterjavanetsocketexception broken pipe at method at at at at at at at at at at at at at weblogicservletinternalchunkoutputwrapperprintchunkoutputwrapper

if i want to stash new created files i stage them first so they are being included in the stash so at the point when i try to stash there are files present in the staged files area i click on stash give it a description and press enter like usual but when there are staged files the focus now jumps to the commit button instead of finishing the stash process the problem is that it may not be noticed at first the enter key is hit again and files are committed

unfortunately jira will not run under the jvm although all appears well the scheduler fails to start quartzscheduler scheduler jiraquartzschedulernonclustered startedexception in thread quartzschedulerthread javalangclasscastexception javautildate at at at at at at at at at at

with confluence some ldap configurations are requiring paged results evidently sun directory ldap doesnt support paged results see also unable to log in to confluence with ldap error code unavailable critical extension due to sun directory server enterprise edition and paged results

cannot start bamboo on windows i tested and did not get the same problem attached is a screenshot showing the error message when the bambooconsolebat is run

this is a ticket to link to the marketplace bug since confluence is not present on marketplace upm on such version gives the following error message noformatthis version of confluence is unknown to the atlassian marketplace it may be too new to appear or it may be some type of nonstandard buildnoformat

problem the show repository browser remote page queries bitbucket server for a list of repositories that the user has permission to however in a large enterprise with thousands of projects and thousands of users you can overwhelm even a multinode data center instance with the number of rest api requests that are made steps to duplicate configure sourcetree to connect to a remote bitbucket server or data center instance that has a significant number of projects on the bitbucket server tail the atlassianbitbucketlog enable debug logging in bitbucket server in sourcetree visit the show repository browser page and then click remote result you will see at least one and maybe two mentions of the rest api pre project that is being called noformat noformat for a small or even medium sized instance this may not be noticed however is a large enterprise with up to users and thousands of projects when more than a few users are accessing the repository browser simultaneously the number of rest api requests that are happening simultaneously can negatively impact the performance of bitbucket server when you further complicate this with using delegated ldap for authentication each of these rest api calls causes a validation to ldap this can also cause thousands of sql update commands to be fired off updating the cwduser table row for the user that is connecting to bitbucket desired behavior a list of possible options not necessarily in any particular order use a different rest api that returns all the users repositories they have access to with one call have the repository browser be paged so that only or projects are queried at a time change the ui so that you get a list of projects first rest api call and the display is updated with the repositories in the selected project rest api call for every repositories in the project make browsing the available repositories optional provide the ability to paste a clone url without having to query every project

note this bug report is for jira service desk cloud using jira service desk server see the corresponding bug report panelhow do i enable stats for cloud

actually thisremote abort repository is in read only modefatal unable to access the requested url returned error

the identifierhash for a session token in atlassian crowd before version could potentially collide with an identifierhash for another user or a user in a different directory this allows remote attackers who can authenticate to crowd or an application using crowd for authentication to gain access to another users session provided they can make their identifier hash collide with another users session identifier hash

after about an hour or two i tried to edit another page in our confluence test instance i got an error when trying to edit the page this seems to be a timeout issue with our jdbc jndi pool used for communicating with the confluence database i attach a combined log of atlassianconfluencelog atlassiansynchronylog and mysql server log the mysql server seems to abort a connection after some time because there was no response to data delivered to the synchrony client i made sure that the mysql server was running there was definitely no connection problem to the server as several page where retrieved and open in confluence is this due to a misconfiguration or a general error in synchrony when i retried editing a page another hour later the problem didnt exist any more might be an issue with reusing a database connection which has been idleclosed when not being used for some time regards janpeter solution we will make the connection pools parameters of synchrony configurable via the system properties basically a confluence administrator can set the following properties synchronydatabasetestconnectiononcheckin default true synchronydatabaseidleconnectiontestperiod defualt seconds or minutes synchronydatabasemaxidletimeexcessconnections default seconds or minutes synchronydatabasemaxidletime default seconds or hours

note this bug report is for confluence server using confluence cloud see the corresponding bug report panelwe believe there may be a small subset of users who will experience this problem when logging in to atlassian answers now that the migration to confluence questions is complete instead of being linked to their own user account they will instead be linked to a different account with a similar name this appears to be related to the way in which atlassian id maps users to local account names we are still investigating the extent of this issue and whether or not we can solve it via a bulk update if you encounter this problem please comment on this issue and we should be able to solve it for your account with a onceoff configuration change

this page indicates that changing the clone prefix would be possiblebut in order to do that admins would need to edit the jira advanced settings which is a restricted function of atlassian ondemand

hello lately when the users create a ticket this ticket is automatically assigned to the projecte lead normally according to the workflow the ticket should be assigned to another user furthermore in the issue details the current assignee is this another user but this user does not see any issue assigned to him when i try to filter the issues then i see that the assignee is the project lead could it be that there is somewhere a contradiction in the assignements

works fine for plan view failed to lazily initialize a collection no session or session was closed

when renaming a page all the links that are pointing to that page by having another title shown get broken example page project plan link on another page to project plan if i rename project plan to project plan details the link gets broken or more specific it simply not update itself with the new name this stands totally in contrast to everything that a knowledge management platform wants to achieve can you have a look at that

i have fisheyecrucible installation that i configured for one perforce repository my duplicate server and it was working fine now i have added a second repository my production server and stopped the first the first shows on the dashboard with a status of stopped but the second does not show at all on the dashboard in the admin page they both appear and the first has a status of enabled stopping and the second has a status of enabled running they have been like this for serveral hours what is going on here

to reproduce this issue create a user with username alerthahahaha create a personal space for this user create a page in the personal space with pagetree andor pagetreesearch macro note that confluence does not work very well with such usernames so you would need to use actions directly when creatingviewing pages in the user space

summary jira rest api has these two similar methods in this documentation get on return information about a user get on information on one of more users when using operation for a user it intermittently fails though we have a success return an empty set of data is returned though the user has data and operation returns correctly this is consistently failing for some users and consistently working for others environment only tested in jira cloud steps to reproduce have a clean jira cloud instance create a user username test make sure it belongs exactly to ‘confluenceusers’ and ‘jirasoftwareusers’ we were able to replicate it this way and since it’s not happening to all users group membership may play a role run this replacing user password and instance accordingly otheruser must have access to the users noformat curl d u otheruserpassword x get h contenttype applicationjson noformat run this theres a search in this one noformat curl d u otheruserpassword x get h contenttype applicationjson noformat expected results both return a ok and user information actual results both return a ok showing the operation is accepted the first operation shows the data from the user but the second returns no data returns ‘’ notes testlink uses ‘’ and may fail because of that if the user selected is affected if you have access to testlink code we’ve been able to fix it removing the ‘search’ from the call just changed the file varwwwtestlinkthirdpartyfaypjirarestjiraphp public function getuser from thisrequestopenconnectthishost usersearchusername username get to thisrequestopenconnectthishost userusername username get if you’re searching data for one user use ‘’ instead

jira uses joda time for time tracking and joda time ships with embedded timezone information when governments change their minds and alter daylight savings times this data is rendered obsolete causing jira to display times incorrectly and jira has no way of updating this information this has recently occurred in argentina which abolished daylight saving time in oracle publish regular timezone data updates to correct this for the core java packages but joda time does not make use of these try this quick test with joda time as shiooed with jira code public void testargentina timezone tz assertfalsedaylight tzindaylighttimenew dateformat format dateformatgetdatetimeinstancedateformatshort dateformatshort formatsettimezonetz systemoutprintlnformatformatnew date datetimezone jtz datetimezonefortimezonetz datetimeformatter jformat datetimeformatforstylesswithzonejtz jformatprintnew orgjodatimedatetime systemoutprintlnjformatprintnew orgjodatimedatetime timezone tz timezonegettimezoneamericaargentinabuenosaires assertfalsedaylight tzindaylighttimenew dateformat format dateformatgetdatetimeinstancedateformatshort dateformatshort formatsettimezonetz systemoutprintlnformatformatnew date datetimezone jtz datetimezonefortimezonetz datetimeformatter jformat datetimeformatforstylesswithzonejtz systemoutprintlnjformatprintnew orgjodatimedatetime code

this needs to be verified after is done