Datasets:

kristmh

/

mongoDB_testset_highest_high

like 0

it would help if mongodb could support queries where you can filter on regex substring of fieldsegsql select from tablewhere substringfieldname abc we use examples of this in our current sql rules environment and need to have this in mongothankskovi

its possible this is because we set searchmeta at runtime and since we build the unionwith subpipeline before the value is set it remains unset on the copied expression context in the subpipeline

the http console ui hits an assert in dbstatssnapshotscpp if the system clock has been set back in time recently including when the clock has been set automatically by ntpd this results in the below messages and an http error from the http consolemongodb nov assertion failure newercreated oldercreated dbstatssnapshotscpp nov assertion failure newercreated oldercreated srcmongodbstatssnapshotscpp usrbinmongod noformatthis is probably the same underlying issue as however i think it should get a second look this makes the console not reliable for automated monitoring also regular users will see the assertion failure in their browser and get no indication whymy plan to avoid hitting this again is to enable the rest api with rest and have our monitoring tool use that insteadmy thought is that since this only seems to affect the dbtopstatus and writelockstatus objects can the failure be caught in there and not prevent the entire http console from loading

tutorial should cover configuring mongodb native ldap for authentication via active directory configuring mongodb for authorization via active directory

going through the deployment wizard we suggest to the user to use cname to avoid having to change connection strings avoid downtime and point to the following documentation the link is confusing as the doc is about config servers and searching for cname gives about the same lines of text as shown in the wizard without adding more we should have a separate doc page on cname and how to obtain them

currently mongoclient database and collection can be mutated through the baseobject attributes readpreference writeconcern uuidsubtype etc this is leading to problems in applications where different threads apply conflicting changes to the same object reports of buggy applications like this are increasingpymongo will make instances of these attributes immutable the existing attributes will become read only uuidsubtype will be removed with its functionality folded into a codecoptions class tagsets will be removed with its functionality folded into read preference and a new api will be provided to create database and collection instances with explicit codecoptions readpreference and writeconcern

expand inactive commits for history tables view is uncluttered by commits that do not run or are not scheduled to run the task in question by default from on the task ui eg for infrequent tasks we would like to be able to hide all of the inactive tasks in the lefthand pane so that we can easily see the redgreen ones ideally this would be a dynamic reloadable toggle

our current packaging system has a bindip parameter that is set to localhost this effectively means that anyone who upgrades will have their sharding and replication broken because they will only be listening on localhost this should be commented out in the configuration when upgradingthis was changed in and should be reflected in the release notes please follow up with the packaging team on the future of this but for now we should warn and guide people appropriately in both the upgrade page and release notes

hello guys may you help me with my problem i installed your product but when i try run command wt into console i see next error wt error while loading shared libraries cannot open shared object file no such file or directory step by step how i installed it git clone gitgithubcomwiredtigerwiredtigergit mv wiredtiger wiredtiger cd wiredtiger sh autogensh configure enablezlib cppflagsiusrlocalinclude ldflagslusrlocalinclude enablepython make make install os linux mint cinnamon

it fails to start with noformat f control failed global initialization unknownerror could not add sasl auxprop plugin mongodbinternalauxprop version mismatch with plugin noformat

many of the links at are deadperhaps this page could be pruned and the list reordered most popular first

mongodb failed to buid due to error boostunorderedsetstdallocator no appropriate default constructor available this issue is caused by revision could you pleaes help to take a look at this thanks in advance some investigate on my side the related code in logicalseesioncachetestcpp stdxpromise hitrefresh auto refreshfuture hitrefreshgetfuture advance time to first refresh point check that refresh happens and that it includes both our records sessionssetrefreshhooklogicalsessionidset sessions logicalsessionidset hitrefreshsetvaluesessionssize return if we change the code as below there is no error reported stdxpromise hitrefresh to stdxpromise hitrefresh

ihac who is trying to run two query conditions in the same match stage from the aggregations component in compass but receives error expected end of input but found both compass version and example ts gte lte i have been able to reproduce the error using the same compass version and mongodb version enterprise workaround which customer is using is to to run the two query conditions in separate match stage through compass have tested using two conditions which does not include isodate and expected results return can you please confirm if this is a bug with compass if so will there be a fix thanks in advance

this bug showed up when i was trying to run a multiversion updatefuzzer task against and on we can set a nonexistent dotted field to an object noformat dbfuzzeraggregate id a b a noformat but on and master this command would somehow result in a noop eg noformat dbfuzzerfind id dbfuzzeraggregate id noformat though we can still set a nonexistent simple field to an object or set a dotted field to a simple nonobject value on master eg noformat dbfuzzeraggregate id a b dbfuzzeraggregate id a a noformat

for environments that cannot afford a risk of data loss or inconsistency ie that would prefer unavailability instead the arbiter could provide a mode where the election of a new primary is allowed only if are absolutely sure that there is an up to date node in the replica set this guarantee can be offered if the primary has been shut down gracefull but not in the case of a crashthe arbiter could handle those situations differently graceful shutdown usual process a new primary gets elected crash or no guarantee that a node is up to date for sure do nothing and wait for a human operator or after a timeout eg hour to tell the arbiter to bypass the security and start the election processthat way we have atotimatic availability when we know there is no risk for the data when there is a risk we give ourself time to try restoring the crashed primary within a define sla period for availability we can always force the process if we can confirm before the end of the sla that there are actually no risks or if there is nothing to recover from the old primary at the end of the sla period for availability the process gets forced automatically

commands to reproduce mongo commands with failure mongo use customernameproductnameenvironment dbacollectionthatshouldhaveashorternameinsertname wintesa new shell mongodump d customernameproductnameenvironment mongodump commands mongodump demonstrating workaround mongo use customernameproductnameenvironment dbacollectionthatshouldhaveashorternamerenamecollectionacollectionthatshouldhaveashor new shell mongodump d customernameproductnameenvironmentlogsmongod vvvvv run mongo commands thu apr query customernameproductnameenvironmentsystemindexes query snapshot true thu apr used cursor thu apr query customernameproductnameenvironmentsystemindexes apr query customernameproductnameenvironmentcmd count acollectionthatshouldhaveashorternameid query thu apr run command customernameproductnameenvironmentcmd count acollectionthatshouldhaveashorternameid query thu apr query customernameproductnameenvironmentcmd command count acollectionthatshouldhaveashorternameid query apr query customernameproductnameenvironmentacollectionthatshouldhaveashorternameid query snapshot true thu apr warning no id index on snapshot query nscustomernameproductnameenvironmentacollectionthatshouldhaveashorternameidthu apr assertion bsonobj size first element eoo mongod mongod mongod mongod mongod mongod mongod mongod mongod mongod mongod mongod mongod mongod threadproxy libsystembdylib pthreadstart libsystembdylib threadstart apr assertion invalid bsonobj size first element eoo nscustomernameproductnameenvironmentacollectionthatshouldhaveashorternameid query query snapshot true thu apr query customernameproductnameenvironmentacollectionthatshouldhaveashorternameid exception apr messagingport recv conn closed apr socketexception remote error socket exception thu apr end connection vvvvv rename collection and run mongodump thu apr connection accepted from apr query admincmd whatsmyuri thu apr run command admincmd whatsmyuri thu apr query admincmd command whatsmyuri apr query admincmd replsetgetstatus forshell thu apr run command admincmd replsetgetstatus forshell thu apr command replsetgetstatus forshell thu apr query admincmd command replsetgetstatus forshell apr query admincmd replsetgetstatus forshell thu apr run command admincmd replsetgetstatus forshell thu apr command replsetgetstatus forshell thu apr query admincmd command replsetgetstatus forshell apr query customernameproductnameenvironmentsystemnamespacesthu apr used cursor thu apr query customernameproductnameenvironmentsystemnamespaces apr query customernameproductnameenvironmentsystemnamespacesthu apr used cursor thu apr query customernameproductnameenvironmentsystemnamespaces apr query customernameproductnameenvironmentsystemnamespacesthu apr used cursor thu apr query customernameproductnameenvironmentsystemnamespaces apr query customernameproductnameenvironmentsystemnamespacesthu apr used cursor thu apr query customernameproductnameenvironmentsystemnamespaces apr query admincmd renamecollection customernameproductnameenvironmentacollectionthatshouldhaveas to customernameproductnameenvironmentacollectionthatshouldhaveashor droptarget undefined thu apr run command admincmd renamecollection customernameproductnameenvironmentacollectionthatshouldhaveas to customernameproductnameenvironmentacollectionthatshouldhaveashor droptarget undefined thu apr command renamecollection customernameproductnameenvironmentacollectionthatshouldhaveas to customernameproductnameenvironmentacollectionthatshouldhaveashor droptarget undefined thu apr new namespace customernameproductnameenvironmentacollectionthatshouldhaveashornew namespace customernameproductnameenvironmentacollectionthatshouldhaveashoridquery admincmd command renamecollection customernameproductnameenvironmentacollectionthatshouldhaveas to customernameproductnameenvironmentacollectionthatshouldhaveashor droptarget undefined apr query admincmd replsetgetstatus forshell thu apr run command admincmd replsetgetstatus forshell thu apr command replsetgetstatus forshell thu apr query admincmd command replsetgetstatus forshell apr connection accepted from apr query admincmd isdbgrid thu apr run command admincmd isdbgrid thu apr query admincmd command isdbgrid apr query customernameproductnameenvironmentsystemnamespacesthu apr used cursor thu apr query customernameproductnameenvironmentsystemnamespaces apr query customernameproductnameenvironmentcmd count systemindexes query thu apr run command customernameproductnameenvironmentcmd count systemindexes query thu apr query customernameproductnameenvironmentcmd command count systemindexes query apr query admincmd availablequeryoptions thu apr run command admincmd availablequeryoptions thu apr query admincmd command availablequeryoptions apr query customernameproductnameenvironmentsystemindexes query snapshot true thu apr used cursor thu apr query customernameproductnameenvironmentsystemindexes apr query customernameproductnameenvironmentcmd count acollectionthatshouldhaveashor query thu apr run command customernameproductnameenvironmentcmd count acollectionthatshouldhaveashor query thu apr query customernameproductnameenvironmentcmd command count acollectionthatshouldhaveashor query apr query customernameproductnameenvironmentacollectionthatshouldhaveashor query snapshot true thu apr used cursor thu apr query customernameproductnameenvironmentacollectionthatshouldhaveashor apr messagingport recv conn closed apr socketexception remote error socket exception thu apr end connection apr flushing mmap took for peter mongodump d customernameproductnameenvironmentconnected to customernameproductnameenvironment to dumpcustomernameproductnameenvironment customernameproductnameenvironmentacollectionthatshouldhaveashortername to dumpcustomernameproductnameenvironmentacollectionthatshouldhaveashorternamebson objects customernameproductnameenvironmentsystemindexes to dumpcustomernameproductnameenvironmentsystemindexesbson objects customernameproductnameenvironmentacollectionthatshouldhaveashorternameid to dumpcustomernameproductnameenvironmentacollectionthatshouldhaveashorternameidbsonassertion nextsafe err invalid bsonobj size first element eoo code peter mongodump d customernameproductnameenvironmentconnected to customernameproductnameenvironment to dumpcustomernameproductnameenvironment customernameproductnameenvironmentsystemindexes to dumpcustomernameproductnameenvironmentsystemindexesbson objects customernameproductnameenvironmentacollectionthatshouldhaveashor to dumpcustomernameproductnameenvironmentacollectionthatshouldhaveashorbson objects

when getlasterror is called with the number represents the total number of nodes that have the write when getlasterror returns null errhowever implementation of returns just replicatedto members omitting the primary which also has the writeaside from it being difficult to later track if application logs have the full writeresult but no information about which node was primary and also has the write it also feels unintuitivei think it would be less confusing to end user if returns a structure with writtento that on success is greaterthan or equal the w value and on failure or timeout writtentolength will be shorter than wheres an examplecodejs updatedexisting true n lastop t i connectionid wtimeout true waited replicatedto err timeout ok

mongod outputs glibc detected mongod double free or corruption prev in the logsthe replicaset seems to have stopped replication for a long time that may be another issue primary started output weird errors double free or corruption process was still alive until we kill it but refuses any connections shell or with other replicasets members

codecodevar mongodb functionerror db dbdropdatabasefunction dbcollectiontestinsert name val result function dbcollectiontestfindone name val functionerror doc consolelogjsonstringifydoc a result of the following code in mongodbcore

we are going to a multimaster setup per sharddetails forthcoming

tracks any microdesign and coding work needed to move the migrate chunk function to the catalog manager

we should add tests with large and complex aggregation pipelines including large and complex expression trees this will help ensure that the visitors are correctly maintaining state as we expect them to be ideally we will also construct test cases that are closer to how we think customers are going to be querying with mongocryptd mixing many language features at once

the copy function for bsonobj could detect corrupt bson and detect undefined behavior earlier for two reasons there is no verification that objsize is below the maximum bson size an attempt to allocate a new buffer based on the size of an unowned bson object can lead to very large allocations by making calls to objsize the copy function may allocate a different amount of memory than it writes if the underlying buffer is unowned this should also be fixed to prevent the possibility of undefined behavior

starting in all database metadata indexes collections must be accessed via getindexes and getcollection helpers rather than reaching directly into the system collections directly the mgo driver will do the right thing automaticallysee

the mongouri testing part of connectionstringtest needs to be updated to reflect dns changes

we encountered this issue since i noticed others were facing the similar issue eg so i waited for stable release hope it can fix the issue but unfortunately after upgrading to we still see this issue there is no obvious pattern about which specific query caused crash but it happens from time to timemongodb run or days then crash especially when query load is heavy every time after crash i need to manually restart the server manual restart is impractical for prod environment can you take a look at my attached log thanks

before the new schema validation experience is released we need to update the welcome carousel with the new visuals and copy

for intensives findandmodify operations which use a service call inc in some cases there is a loss and this effect is mistery no memory leak detected memory and cpu working correctly test also working on other osexample for nodejs db already connected for databasebegindbcollectiontestsavea inc test approximately to operations there is a loss on the errorthis is math error maybe in engine ps spidermonkey engine dont have this bug im already test it

i was trying to use an aggregation pipeline in the following order match sort in my case match step returned documents sort step sorts data but they all have the same value in the sorted column skip step skips first documents limit step limits the result to documents i expected to see last documents from the match result but the query returned and documents from the match result if i set a limit to then everything works as expected i receive last documents the other way to get a correct result is to add an id as the last sort column in sort step or exclude sort step at all if i use simple chaining like findsortskiplimit it works correct too

mongodb introduces compatibility changes regarding the ability of a multikey index to be used in sorting to summarize in mongodb a version or earlier multikey index mongodb and earlier cannot be used to sort a version multikey index mongodb can be used to sort provided that the sort field is not an array given the performance implications and that the indexes already contain this information collectiongetindexes should display the index metadata identifying if an index is multikey if a v show the path level multikey data for any multikey indexes

log message says rsm monitoring host in expedited mode until we detect a primary this only shows up once when rsm turns expedited mode on until it detects a primary and turns off expedited mode in the case when there is always a primary and expedited mode is being repeatedly turned on the log message will show up multiple times in a row and can be easily misinterpreted as rsm thinks that there is no primary the expedited mode also sounded like some kind of feature and has confused a lot of engineers i propose that we remove the mention of until we detect a primary reword expedited mode into something like rsm temporarily going to refresh sooner log in question

based on messages produced by tests in evergreen it looks like these directories may not have been scrubbed yet to complete the conversion let me know if these are already covered by another ticket youre having trouble finding the log statements that need to be converted or if these belong to a different team srcmongoscriptingmozjs srcmongoshell

each mongod and mongos process keeps an account of its activity in a log file to view logs click clusters then the mongodb logs tab refresh your browser to view new entries for information on reading mongodb logs see log messages in the mongodb manual you can optionally disable log collection for your group atlas will continue to display previously collected log entries but will cease to collect new log data to disable log collection click settings then group settings set collect logs for all hosts to off you actually need to click the individual hosts and then go to the more tab

in the doccodeborderstylesolid dbsystemprofilefindts thu jan est info query testcmd query profile millis the shellcodeborderstylesolid dbsystemprofilefind ts op query ns testsystemnamespaces query nscanned nreturned responselength millis client user ts op command ns testcmd command create capped null size null max null ntoreturn responselength millis client user codedifference in info section and opnscommandquery

attached program crash with core dump backtrace noformat gdb bt in wtstructsizev session sizep fmt ap at in wtcursorsetkeyv cursor flags ap at in wtcurtablesetkey cursor at in curjoinnext cursor at in main at noformat it happens for but for development branch works well so looks like this issue already fixed

the ensureindex changes the output which is badnot sure what should be correct but having an index shouldnt change outputdbfoodropdbfoosave a dbfoosave a dbfoosave a dbfooensureindex ax dbfoofindsort ax foreach printjson

as we have reimplemented all the connection logic the existing driver needs to be refactored to use the new implementationbackwards compatibility will be important but we will not sacrifice maintainability for instance mongoconnection is public but it is highly unlikely anyone is using it in this case it will probably get deleted however mongoserver and mongoserverinstance are public and we know of a number of people utilizing them and therefore we will reimplement mongoserver and mongoserverinstance by wrapping the core concepts icluster and iserver respectively

the mongoldap validation tool should be tested in the evergreen ci loop the jstest should run mongoldap several times with different configurations against the ldap server in aws it should assert the expected return codes for each configuration

when we sweep the lookaside table we can only start removing a block at a full record if we start at a modify update that update will be lost when the record is read

description hi team can we please update to change mongodb to mongodb products in the following sentence from if you believe you have discovered a vulnerability in mongodb or have experienced a security incident related to mongodb please report the issue to aid in its resolution change to if you believe you have discovered a vulnerability in mongodb products or have experienced a security incident related to mongodb products please report the issue to aid in its resolution cc please see the copy change scope of changes impact to other docs mvp work and date resources scope or design docs invision etc

a quick grep for cursortype reveals that its only mentioned in optionsfindx never in any of the implementation files so we just dont propagate it and theres no way to ask for a tailable or await cursor right now mentioned in the mongodbdev group at

when a query run against a wiredtigerenabled mongod yields or is saved for a getmore the querys storagelayer snapshot is dropped and later replaced by a new snapshot for queries that perform multiple index lookups for a single document this creates an opportunity for index key data saved by the query stage tree to be stale by the time the new snapshot is created as a result queries can return documents that never match the given predicate if they partially match the predicate in an old snapshot and partially match the predicate in a new shapshot for example the below script shows a query with predicate a b returning a document that matches predicates a and b at separate times but never at the same timethis issue does not affect mongod configured with the storage engine as the write invalidation framework forces queries to perform a full predicate reevaluation on documents that are mutated over the course of the querys lifetimeto reproduce run the following shell script against a mongod configured with the wiredtiger storage enginecodejsdbfoodropfor var i assertwriteokdbfooinsertid i a b a b add artificial score boost to andsorted plansassertcommandworkeddbadmincommandsetparameter internalqueryforceintersectionplans true increase frequency of yieldingassertcommandworkeddbadmincommandsetparameter internalqueryexecyielditerations in a parallel shell continuously flop the a and b values of the id documentstartparallelshell while true assertwriteokdbfooupdateid set a b assertwriteokdbfooupdateid set a b while true run an andsorted intersection query on a predicate that matches no documents this below assertion fails when the query incorrectly returns the id document dbfoofinda b id a b

evergreen should either retry or fail instead it left the host up in the building state noformat describe instances response has no reservations githubcomevergreencievergreenunitscreatehostjobcreatehost githubcomevergreencievergreenunitscreatehostjobrun githubcomevergreencievergreenvendorgithubcommongodbamboypoolrunjob githubcomevergreencievergreenvendorgithubcommongodbamboypoolexecutejob githubcomevergreencievergreenvendorgithubcommongodbamboypoolabortablepoolrunjob githubcomevergreencievergreenvendorgithubcommongodbamboypoolabortablepoolworker runtimegoexit noformat

since ops manager was already released listcommands must always return ismaster in all subsequent server versions of once cloud completes downstream work for we can remove this from listcommands

i just received a ticket for someone asking for a feature that is already on the settings page in the process i noticed we do not have an explicit section documenting the various options

can not edit this page error when i try to do so would like to add a new delphi driver

please see issue

we should either use the correct major version number or change the hardcoded string to linux sans version

should overview the major changes and new features in the driver

on the task queue page in the distro at this moment there are queued tasks i clicked on the replicasetswt task and on the task page it shows in queue but on the queue page it shows as i refreshed both pages to make sure there wasnt stale data on one or the other i saw the same thing with the task on the queue page replicasetsauth it shows in queue even though its on the queue page urls i was hitting and screenshots attached for the three pages

for the community build

create a table with one index and dump it the table shows up but the index does not current tests test are either too simple tables without indices or too complex tables which have both column groups and indices the existence of a column group makes the problem go away

i click on current and get an options pane which allows me to select or for the version whichever version i select i still get the docs

has an index size reference that doesnt actually link anywherenoformatwhich include the full content up to the maximum index size ofthe subdocument in the index instead indexes on embedded fieldsnoformat

as a follow up to given the impact this issue may have to the customers upgrading from release it would be nice if the write concern article for mongodb had an attention box saying that fireandforget no longer works for sharded clusters etc with a link to the compatibility changes

task logs failed on the linux no journal build variant because it passes nojournal to resmokepy which will pass nojournal to all started mongod processes noformat starting mongod on port mongod setparameter dbpath replset configrs nojournal nopreallocj storageengine wiredtiger configsvr port mongod started on port with pid starting mongod on port mongod setparameter dbpath replset configrs nojournal nopreallocj storageengine wiredtiger configsvr port mongod started on port with pid starting mongod on port mongod setparameter dbpath replset configrs nojournal nopreallocj storageengine wiredtiger configsvr port mongod started on port with pid f control failed global initialization badvalue nojournal is not allowed when configsvr is specified f control failed global initialization badvalue nojournal is not allowed when configsvr is specified f control failed global initialization badvalue nojournal is not allowed when configsvr is specified noformat necessary after

for immediate release version released fixes issue in which the automation agent crashes when attempting to load an invalid certificate bundle

we specifically johnmorales have observed an issue where the same metrics pings seem to be posted twice by the mongod and at odd intervals specifically in this case every minute at seconds and seconds past code debug metrics metrics request received from rid debug metrics metrics request received from rid debug metrics metrics request received from rid debug metrics metrics request received from rid debug metrics metrics request received from rid debug metrics metrics request received from rid code our hunch is that either there are two post threads running slightly out of sync or two timers triggering the posts this seems to happen indefinitely on the affected mongod and never fixes itself this issue has so far only happened once and we havent been able to reproduce it

our waterfall shows nondeterministic failures on macos where the test execution times out after lets investigate these to see if something is going wrong or if the timeout just needs to be increased edit in the patch build for i also noticed this for some windows tasks patch build

ive run into this issue today on ubuntu using mongod ce sudo service mongod start fails looking into the log varlogmongodbmongodlog shows that it failed to start because another process is listening socket in use error code doing a ps wuax grep mongo i can see that a process of mongod is already running as mongod bindipall user killing it with kill pid doesnt do anything the process respawns using pkill or htop to send different kill commands also fail the process stays thererespawns with new pid terminating it with sudo service mongod stop doesnt do anything it still stays running connecting to it via mongo cli simply hangsup at the connecting part this happened without any abnormal termination or closure of the platform short of uninstalling and reinstalling i have no other way to use mongod at this point

hi we are getting an issue in user creation call of rubydriver to be exact the issue occurs in this particular line error thrown noformat error undefined method for noformat stack trace noformat i info – creating user for db i info – backtrace block in updateclustertime synchronize updateclustertime login block in authenticate handleauthfailure authenticate connect ensureconnected write deliver block in dispatch publishcommand dispatch block in dispatchmessage withconnection withconnection dispatchmessage execute execute block in create withsession withsession create noformat

a mongodump with compression to an output directory not archive rapidly consumes all available ram and is terminated by oomkiller eg with data size mongodump reached ± rss before being terminated a few times there was a stacktrace which ive attached mostly its oomkiller note without the gzip option the dump completes successfully mongodump gzip archivevarbackupsmongodbtest completes successfully noformat out of memory kill process mongodump score or sacrifice child killed process mongodump noformat noformat with time v mongodump gzip out command terminated by signal user time seconds system time seconds percent of cpu this job got elapsed wall clock time hss or mss average shared text size kbytes average unshared data size kbytes average stack size kbytes average total size kbytes maximum resident set size kbytes average resident set size kbytes major requiring io page faults minor reclaiming a frame page faults voluntary context switches involuntary context switches swaps file system inputs file system outputs socket messages sent socket messages received signals delivered page size bytes exit status noformat the exact command line arguments being supplied noformat mongodump ssl sslallowinvalidcertificates host hiddensecondaryexamplecom port username backup password redacted authenticationdatabase admin out varbackupsmongodbtest gzip noformat

code filter options query new mongodbdriverqueryfilter options vardumpoptions code the sort option is changed to a stdclass instance code options filter options query new mongodbdriverqueryfilter options vardumpoptions code the above results in noformat script freeing bytes total memory leaks detected noformat dumping the sort option from after a second query construction results in its value being displayed as which indicates some corruption lastly its possible to invoke a segfault by executing one of these queries see segfaultphp

helloas the order of criteria fields matters i was wondering why not always sort the fields by keyname before creating indexes and searchingiedbmycollectionensureindexb c a internally could be replaced by dbmycollectionensureindexa b c dbmycollectionfindc c b b a a internally could be replaced by dbmycollectionfinda a b b c ci think it could help to match the indexes easier especially in cases we add the criteria fields based on user filter selections complex viewsdoes that make sensethanks

see initial comment for the description

the image hereseems to show a single machine in each region for each shard tagged jp for example best to have several replicaset members deployed in japan with high replicaset priority and then deprioritize the rs members out of japan then if the primary goes down another machine in japan will take over the write load from app servers in japan same obviously for the other regions besides japanas currently drawn the cluster is deployed so that if a machine goes down in japan a secondary in ny or london will take over as primary and applications will send writes across the globe

relevant to mongofiles mongoimportmongoexport mongorestoremongodumpalso storage related options should error out intelligently and not listed in the help directoryperdb journalshould also cleanup optionsgo while were at this

the replica set secondary crashed during initial sync in the build index step while building the index on a mil docs large collection code i index build i index build i index build i network end connection connections now open i network connection accepted from connections now open i network end connection connections now open i network connection accepted from connections now open i access successfully authenticated as principal system on local i access successfully authenticated as principal system on local i index build i index build e storage wiredtiger cursornext encountered an illegal file format or internal value e storage wiredtiger cursornext the process must exit and restart wtpanic wiredtiger library panic i fatal assertion i control begin backtrace backtraceprocessinfo mongodbversion gitversion uname sysname linux release version smp fri dec est machine somap mongod mongod mongodwterr mongodwtpanic mongodwtbtcurnext mongod mongod mongod mongod end backtrace i aborting after fassert failure code

we are seeing periodic failures from testcheckpoint at the moment due to a bug in its use of the new timestamp api there is a ticket open to fix testcheckpoint but its blocked waiting for some api changes in timestamp support this ticket captures the work to temporarily disable running testcheckpoint with timestamps in order to avoid spurious test failures

when i backup my mongodb deploymengan error says that backup file is too large to be converted to an int error log exception caught at the root level javalangarithmeticexception the byte count is too large to be converted to an int

isnt the connections between agents and ops manager section superseded by a separate page that details how to set that up this content seems out of date allisons to do notes update the backup monitoring agent configure agent for ssl tutorials to link to the configure ssl connections to ops manager tutorial rather than talking about using a proxy for ssl remove weird link to the mongodb ssl docs add links to the monitoringbackup agent configure tutorials to the configure ssl connections to ops manager tutorial in a seealso change configure ssl connections to ops manager tutorial title to something a bit clearer and add some conceptual stuff about what exactly youre doing in that tutorial

it looks like the code throws new userexception while most mongo code like uasserted throws userexception and the higher level trycatch block for each operation cannot handle the formertestcodec dbccdropcensureindex csave loc should uassert instead of causing process to terminateccount loc foo codeoutputcodemongod help for help and startup optionsthu nov mongodb starting dbpathdatadb hostfedorathu nov thu nov note this is a development version of mongodbthu nov not recommended for productionthu nov thu nov db version pdfile version nov git version nov build info linux fedora smp mon may utc nov options thu nov journal dirdatadbjournalthu nov recover no journal files present no recovery neededthu nov waiting for connections on port nov admin web console waiting for connections on port nov connection accepted from connection now openthu nov cmd drop testcthu nov build index testc id thu nov build index done scanned total records secsthu nov info creating collection testc on add indexthu nov build index testc loc x thu nov build index done scanned total records secsthu nov end connection connections now openthu nov connection accepted from connection now openthu nov cmd drop testcthu nov build index testc id thu nov build index done scanned total records secsthu nov info creating collection testc on add indexthu nov build index testc loc x thu nov build index done scanned total records secsthu nov uncaught exception terminatingthu nov dbexitthu nov shutdown going to close listening socketsthu nov closing listening socket nov closing listening socket nov closing listening socket nov removing socket file nov shutdown going to flush diaglogthu nov shutdown going to close socketsthu nov shutdown waiting for fs preallocatorthu nov shutdown lock for final committhu nov shutdown final committhu nov shutdown closing all filesthu nov closeallfiles finishedthu nov journalcleanupthu nov removejournalfilesthu nov now exitingthu nov dbexit exiting immediatelycode

mongodb weak password controls can result in unauthorized access and modification leakage of data this is a risk factor highlighted by irmc as a part of sox compliance mongostat version mongostat version git version

we need a stronger warning about the use of encryption for createuser warning please note that by default createuser sends a password in plaintext to mongodb which can compromise it on insecure networks use of ssltls to protect the password in transit is strongly recommended many environments require encrypted communications for authentication there already is an important box and also a warning box above the encryption section perhaps to avoid too many boxes competing for attention we could we remove the important box it says an error will be thrown so users will be made aware of this by the error and downgrade the warning box about usability to an important box leaving only the new createuser warning box it is very important we warn people to encrypt this password to avoid security breaches

on for on the seconds step create a file for mongodb command for debian is using placeholders instead of version noformatecho deb stretchmongodborgpackagebranch main sudo tee etcaptsourceslistdmongodborgversionlistnoformat should be noformatecho deb main sudo tee

the migratefromstatus data structure is largely synchronized by the lock on the collection containing the chunk that is currently migrating this implementation assumes that modifications of a collection exclusively lock that collection which is no longer the case for documentlevel locking storage engines

during windows crash handling in the unhandledexceptionfilter we should terminate the process instead of call exit this avoids running the subset of the c atexit handlers that run during exit and os clean exit behavior like dllthreaddetach

will be the source for any additional details you need please refer to the help ticket to get started with grooming

hi i have noticed that the cloud manager docs here do not mention that you should suspend automation before removing replica set members from a managed deployment this could cause issues for customers and should be highlighted clearly in the docs thanks barry

when the createindexes command is replicated to a secondary if the secondary sees an argument that it doesnt understand it should return an error and thus abort rather than ignoring the argument this impacts indexes only which are new with mongodb

at least three times on different versions we have gotten into this situation attached are the database files with the corrupt datathe general flow is that the leaderboard collections get created and dropped based on time and other variables they last for about is a group that is run on these tables every minutesfor examplequery command group ns leadeboardxxxxxxxxxxx key value true cond null reduce functiondocprev prevnumentries initial numentries generate some data on the rank here is a sample of magnitude of the problem collections with only unique names echo show collections mongo sort u wc l show collections mongo wc l collections will show partial

the docs for backup a sharded cluster with database dumps are relying on outdated parameters for mongodump specifically journal and dbpath are no longer supported see here

kfir bloch kfirondev dear mongodb why the documentation on java driver contains deprecated function please review if there are any portions in the driver that have outdated comments thank you

there are bugs in the client side encryption doc java driver the second example does not compile it should be noformat var datakeyid clientencryptioncreatedatakeylocal new datakeyoptions noformat instead of noformat var datakeyid keyvaultcreatedatakeylocal new datakeyoptions noformat and noformat var autoencryptionsettings autoencryptionsettingsbuilder keyvaultnamespacekeyvaultnamespace kmsproviderskmsproviders schemamapmapof noformat instead of noformat var autoencryptionsettings autoencryptionsettingsbuilder keyvaultnamespacekeyvaultnamespace kmsproviderskmsproviders namespacetolocalschemadocumentmapmapof noformat i am talking about this page also it is only working in java java we should add a comment that you need java or provide a java example i am seeing lot’s of developers struggling with our examples because they are not as skilled as you might expect them to be here is my example that works in java

a customers server has stopped working after upgrading from to attempting to start the mongod process with the configuration file results in the following log codejava i control mongodb starting dbpathvarlibmongodb hostrforums i control db version i control git version i control openssl version openssl mar i control allocator tcmalloc i control modules none i control build environment i control distmod i control distarch i control targetarch i control options config etcmongodconf net bindip port security authorization enabled storage dbpath varlibmongodb journal enabled true systemlog destination file logappend true path varlogmongodbmongodlog w detected unclean shutdown varlibmongodbmongodlock is not empty i detected data files in varlibmongodb created by the wiredtiger storage engine so setting the active storage engine to wiredtiger w storage recovering data from the last clean checkpoint i storage i storage warning using the xfs filesystem is strongly recommended with the wiredtiger storage engine i storage see i storage wiredtigeropen config e storage wiredtiger error txnrecover read checksum error for block at offset block header checksum of doesnt match expected checksum of e storage wiredtiger error txnrecover encountered an illegal file format or internal value e storage wiredtiger error txnrecover the process must exit and restart wtpanic wiredtiger library panic i fatal assertion at srcmongodbstoragewiredtigerwiredtigerutilcpp i aborting after fassert failure f got signal aborted begin backtrace backtraceprocessinfo mongodbversion gitversion compiledmodules uname sysname linux release version smp tue sep utc machine somap mongod mongod mongod mongodwteventv mongodwterr mongodwtpanic mongodwtblockextlistread mongodwtblockextlistreadavail mongodwtblockcheckpointload mongod mongodwtbtreeopen mongodwtconnbtreeopen mongodwtsessiongetbtree mongodwtsessiongetbtree mongodwtsessiongetbtreeckpt mongodwtcurfileopen mongod mongod mongod mongodwtlogscan mongodwttxnrecover mongodwtconnectionworkers mongodwiredtigeropen mongod mongod mongodmain mongodstart end backtrace code files in the directory codejava rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb oct rwrr mongodb mongodb aug rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb sep drwxrxrx mongodb mongodb sep diagnosticdata rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb jul rwrr mongodb mongodb aug rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb aug rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb jul rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb jul rwrr mongodb mongodb sep rwrr mongodb mongodb jun rwrr mongodb mongodb jul rwrr mongodb mongodb jul rwrr mongodb mongodb jul rwrr mongodb mongodb jul rwrr mongodb mongodb jul rwrr mongodb mongodb sep rwrr mongodb mongodb jul rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb aug rwrr mongodb mongodb sep rwrr mongodb mongodb sep rwrr mongodb mongodb sep drwxrxrx mongodb mongodb oct journal rwrr mongodb mongodb aug mdbcatalogwt rwrr mongodb mongodb aug mongodlock rwrr mongodb mongodb sep sizestorerwt rwrr mongodb mongodb jun storagebson rwrr mongodb mongodb jun wiredtiger rwrr mongodb mongodb aug wiredtigerlaswt rwrr mongodb mongodb jun wiredtigerlock rwrr mongodb mongodb sep wiredtigerturtle rwrr mongodb mongodb oct wiredtigerwt code any help would be appreciated to get the database back and running thanks

、there are same programs in different pc server，and the config parameters of one program is below： mongoordersocketkeepalivetrue mongoorderwriteconcernmajority 、when there are clients visit the same programs at the same timein order to get data the mongos get below connections in totalbut one secondary of one shard get more than collectionsi think there has some problems about the connections from mongos to shards 、i thried to modify parameter in one mongos or one secondary of shardbut it doesnt workso i want to know how to control the connections from mongos to shards 、the file questionstaskexecutorpoollog is the log of the mongosthe file shardlog is the log of shardthe file metrics are diagnostic data thank you very much

currently redact is more like um remove when conditions are true it strips all the content away theres not even a trace that it existed which is slightly different than redact a very useful variation of this would implement the following actions to replace info instead of stripping it away value a specific value lorem give a certain keyword arg ssn phone address email produce a random but constructappropriate value very important for testing where the value has to bubble up to the business logic in a form that doesnt blow up the business logic random a totally random value appropriate for the type hash a salted hash representation of the content guaranteed to produce the same digest for the same field value this is very important for consumers who need to test with concealed data but in pulling it into their code logic need to keep the proper distribution of values the salt is scoped to the collection as long as at least one record is in the collection the salt lives for that collection

message was lost in but its needed for cloud operators to know the pid of the daemonized process

our mongodb service went down recently with the following error in the logmapviewoffileex for failed with not enough storage is available to process this command file size is in memorymappedfilemapfri dec xxyy fatal assertion after fassert failurewe have a batch job that clears out the collection every night and we do not seem to have had a spike in usage although the datafiles were cleared up by the team that got the service running again so cant be surethe service was in the bad state after crashing and could not simply be restartedthis error seems to have been reported in previous tickets eg however it is recorded as fixed in version

enable the shellapi sharding tests by setting up a sharded cluster using mlaunch for those tests specifically

mongosim based microbenchmarks that update a hot document ie all updating same document show significant performance degradation today threads trying to update the same document throughput for the two is a table scan based and is index scan based last two numbers are for builds from and respectively first three from last week

destroying a mongoc client client pool performs io by sending endsessions to the server destroying a cursor or change stream can lead to mongoc sending killcursors destroying a session can lead to mongoc sending aborttransaction all of these methods are not really suitable to be operations as they dont use connection s instead we should introduce an execute method on the operationexecutor that allows running these asynchronously in the thread pool

tests hang when stubbing out timenow on hi i tried upgrading from to but my rspec tests hang if i have this in my test code letnow timenow before each do allowtimeto receivenowandreturnnow end code if i uncomment out the allow statement the tests run i have not yet been able to determine where it hangs heres the log output when i run the tests code d debug mongodb appboydotcomtestlistcollections started cursor nameonlytrue filternamenotsystem lsidid d debug mongodb appboydotcomtestlistcollections succeeded d debug mongodb event d debug mongodb topology type unknown initializing d debug mongodb event d debug mongodb topology type unknown changed to type unknown d debug mongodb event d debug mongodb server initializing d debug mongodb event newtrue utc readonlyfalse d debug mongodb server description for changed from unknown to standalone d debug mongodb event d debug mongodb topology type unknown changed to type single then it hangs here no logs after this point code

in this article there two examples of security configuration block security options enableencryption … key management configuration options enableencryption … one of them comes with comma at the end of the line and one without which of them should i use please specify and unify this aspect

test that session storage statistics are part of a slow operation report create a doc ticket

mongodb allows keys that contain dots quotefield names can contain dots ie and dollar signs ie quote source however the node driver prohibits them and throws the following error codejava error key mykeywithdots must not contain at serializeinto at serializeobject at serializeinto at serializeobject at serializeinto at serializeobject at serializeinto at serializeobject at serializeinto at bsonserialize at querytobin at serializecommands at poolwrite at executewrite at wireprotocolinsert at serverinsert code the operation was a simple collectioninsertone

here is a coding error it is not related to the database seem to be a driver only error my server environment works but my rspec tests dont work see backtracepng for the error its all about the mongo and mongoid gems i tried with mongoid and although the error message is a bit different with see initpng the initialization is correct at least that part the read value which is a hash since the read mode primary is taken out of the hash see readmodepng finally the options read value is no longer a hash of course it was taken out of it in step this causes the exception we see in step the validate call errors out

the primer layout largely duplicates the mongodb theme layout with some modifications at the bottom to avoid divergence we should inherit the mongodb layout and plop in modifications as needed as we do in the other projects

there are normal tutorials for things like how to restore a cluster this is not covered with the mms on prem docs rather only partial things are covered like how to restore a secondary or standalonenot all customers can make the leap to infer how this relates to restoring an entire cluster our docs need to cover how to restore an entire sharded clusterthis should include something about the balancer being disabled every minutes so that everyone knows the safe things to restore

when using the blanks to fill in the connection information ssh tunnel info is not saved properly and does not display when looking at the favorite additionally the mongo connection port is changed to a seemingly random number over this effectively breaks the favorites functionality for anyone using ssh tunnels i am selecting save changes before i connect ive attached screenshots beforeconnect the favorites connection you can see im connecting over the standard port beforetunnel the tunnel info ive blacked out username and identity file but you can see its connecting over the standard ssh port connect when i connect to the instance you can see that the port info has already been munged the ssh port is showing as and the instance port as a seemingly random number afterconnect the favorite now has yet a different instance port that its trying to connect to aftertunnel all ssh tunnel info has been lost