Datasets:

kristmh

/

clean_MongoDB_balanced

like 0

code is used in two places in the code for different casessrcmongoclientdbclientrscpp line string errmsg try if newconnconnect h errmsg throw dbexception errmsg log name endl catch dbexception e warning name causedby e endl codesrcmongodbdatabasecpp line todo protect against getting sprayed with requests for different db names that dne that would make the dbs map very large not clear what to do to handle though perhaps just log it which is what we do here with the bool cant lockiswritelockedns if loglevel msize cant debugbuild log opening db pathdbpathpath dbname endl cant open database in a read lock if db was just closed consider retrying the query might otherwise indicate an internal error cantcodeit seems bad to reuse an error code but also bad to change one that a customer might be using the one in databasecpp suggests in its text that it cant be interpreted unambiguously so maybe its the one that should be changedit seems dangerous to use error codes directly in throw dbexception statements because our automated tools for preventing duplicate ids will not detect them

while running a ycsb workload on instances the connections between the primary and the secondaries timed out and this lead to one of the secondaries taking over as the primary

broken link on links at bottom of list survey of admin uis for mongodb link goes to this returns a error

here is bson from one of our corrupted records our object structure specified only broken key how it should becode offers lastoffer date list codehow it actually is in bson abovecode offers lastoffer date list list can be any idx codeyes same keys in one object with different values in bson first is array and second is hash due different driver realisations mongo console returns first version and ruby driver returns second hashi tried to run dbcollectionvalidatetrue but mongo didt show errorshere is our usage that can cause itwe have different scripts both use ruby one script makes push to offerslist and second script do set on array member offerslistidx i happens in different sessions im not calling set and push in same updatefor some reason sometimes mongo thinks that array is empty key and when i call set offerslistidx it creates new version of this field which looks like offers list idx so it creates new hash instead of modifying arrayfrankly i cant really understand if its driver bson gem or mongo issue

currently we store bytes in the task document for generate tasks instead of bytes we should store strings so that someone reading the task document in the database can examine the json

if the migration is the first shardingaware operation that the recipient shard is participating in then the recvchunkstart command it receives will make it shardingaware and load the chunk data for the chunk being migrated for the first time if the collection was only recently created and sharded theres a risk that it could load out of date data for the collection causing it to see no chunks for this collection which causes the migration to fail with an obscure data transfer error if the source shard included the config server optime when it sent recvchunkstart then thered be no risk of the recipient shard reading stale data

in the new ui we should switch all displays to convert to time zones in datefns or whatever date library we switch to that way we can stop maintaining our own list

diff with a simple fix built in version git aclientdbclienth bclientdbclienthindex aclientdbclienth bclientdbclienth namespace mongo class dbclientbaseextern resetshardversioncb connectionstring handles parsing different ways to connect to mongo and determining method samples namespace mongo dbclientbase writeconcern wnormal dbclientbase resetshardversioncb this writeconcern getwriteconcern const return writeconcern void setwriteconcern writeconcern w writeconcern w code

it was removed in

backup agent released critical update for users running mongodb with authentication enabled backup agent now includes systemversion and systemrole collections from the admin database in the initial sync

we have requests from customers on how to verify that the config servers are healthy one way is to put them in readonly mode and run however it is not well documented here is a list of few things the doc should address putting the config servers in readonly mode getting a hash for the config dbs list of collections to consider and the ones to not consider when comparing values also adding contents to the dbhash page would be good

after upgrading to the latest c driver the dotnet core web application using the mongo driver and running on azure web apps for containers linux docker image cannot start up any more i am getting the exception below codejava application startup exception systemtimeoutexception a timeout occured after selecting a server using compositeserverselector selectors mongodbdrivermongoclientaresessionssupportedserverselector latencylimitingserverselector allowedlatencyrange client view of cluster state is clusterid connectionmode automatic type unknown state disconnected servers serverid clusterid endpoint endpoint state disconnected type unknown heartbeatexception mongodbdrivermongoconnectionexception an exception occurred while opening a connection to the server systemplatformnotsupportedexception socketiocontrol handles windowsspecific control codes and is not supported on this platform at systemnetsocketssocketpalwindowsioctlsafeclosesocket handle iocontrolcode byte optioninvalue byte optionoutvalue optionlength at iocontrolcode byte optioninvalue byte optionoutvalue at mongodbdrivercoreconnectionstcpstreamfactorycreatesocketendpoint endpoint at mongodbdrivercoreconnectionstcpstreamfactorycreatestreamasyncendpoint endpoint cancellationtoken cancellationtoken at mongodbdrivercoreconnectionsbinaryconnectionopenhelperasynccancellationtoken cancellationtoken end of inner exception stack trace at mongodbdrivercoreconnectionsbinaryconnectionopenhelperasynccancellationtoken cancellationtoken at mongodbdrivercoreserversservermonitorheartbeatasynccancellationtoken cancellationtoken serverid clusterid endpoint endpoint state disconnected type unknown heartbeatexception mongodbdrivermongoconnectionexception an exception occurred while opening a connection to the server systemplatformnotsupportedexception socketiocontrol handles windowsspecific control codes and is not supported on this platform at systemnetsocketssocketpalwindowsioctlsafeclosesocket handle iocontrolcode byte optioninvalue byte optionoutvalue optionlength at iocontrolcode byte optioninvalue byte optionoutvalue at mongodbdrivercoreconnectionstcpstreamfactorycreatesocketendpoint endpoint at mongodbdrivercoreconnectionstcpstreamfactorycreatestreamasyncendpoint endpoint cancellationtoken cancellationtoken at mongodbdrivercoreconnectionsbinaryconnectionopenhelperasynccancellationtoken cancellationtoken end of inner exception stack trace at mongodbdrivercoreconnectionsbinaryconnectionopenhelperasynccancellationtoken cancellationtoken at mongodbdrivercoreserversservermonitorheartbeatasynccancellationtoken cancellationtoken at mongodbdrivercoreclustersclusterthrowtimeoutexceptioniserverselector selector clusterdescription description at mongodbdrivercoreclustersclusterwaitfordescriptionchangedhelperhandlecompletedtasktask completedtask at mongodbdrivercoreclustersclusterwaitfordescriptionchangediserverselector selector clusterdescription description task descriptionchangedtask timespan timeout cancellationtoken cancellationtoken at mongodbdrivercoreclustersclusterselectserveriserverselector selector cancellationtoken cancellationtoken at mongodbdrivermongoclientaresessionssupportedafterserverselectioncancellationtoken cancellationtoken at mongodbdrivermongoclientaresessionssupportedcancellationtoken cancellationtoken at mongodbdriveroperationexecutorstartimplicitsessioncancellationtoken cancellationtoken at func cancellationtoken cancellationtoken at filter options cancellationtoken cancellationtoken at filter options cancellationtoken cancellationtoken at cancellationtoken at source cancellationtoken cancellationtoken code

this smoke test will run as part of our test suite in evergreen it will start the runner and the web service start a container for the agent insert some data into the database run a project config containing a minimal example of each command

diff markup appears on the live site i see the following in the browser head expire at a certain clock time seealso currentdate operatorexpire documents at a certain clock time and edits

the current serves two purposes testing the functionality of btree logic and acting as a general storage engine these purposes are placing divergent requirements upon the record store so it is necessary to create a new simple record store for testing btree logic

in the listener if no ip addresses are supplied we bind to in a testing environment we only need to bind on localhost binding to invokes the os x firewall and pops up dialogue boxes about it for every process that resmokepy starts i believe that this can be fixed by setting the bindip startup option to when starting mongod and mongos

hi im upgrading a rails app with mongoid but after the upgrade i find that options client or method field names are not allowed this is the conclusion i arrived to after looking at mongoid code and seeing this exception nameerror undefined method options for classscheduledreport mongoid origin is there any way to define a field foo that maps to options or whatever field in the database thanks

this is failing consistently four times so far on windows and windows failed function repltestgetmaster reload who the current slaves are var slaves repltestlivenodesslaves var len slaveslength var ready true forvar ilen i var ismaster slavesgetdbadminruncommandismaster var arbiter ismaster undefined false ismaster ready ready ismaster arbiter return ready msgawaiting secondarieserror printing stack trace at printstacktrace at doassert at functionassertsoon at replsettestawaitsecondarynodes at replsettestinitiate at known passing commit

if we are running a standalone mongod executing the following commands from a shell will cause an invariant failure codejs use test dbtestruncommandfindtest readconcern levelmajority afterclustertime new code there is an issue with the way read concern options are handled in commandrun in dbcommandsdbcommandscpp the waitforreadconcern function will try to execute makenoopwriteifneeded if the read concern level is majority and there is an afterclustertime field which will in turn try to run getmylastappliedoptime which will cause this invariant failure if we are not running as a replica set to fix this bug i think the simplest approach may be to bypass any handling of readconcern options in commandrun if we know that we are not running as a replica set or masterslave node

its currently possible to create a patch with a single execution task of a display task create a tag which only contains part of an execution tag that only activates the tagged tasks if a tag includes or more execution tasks from a display the task every task from the display task should be included in all other situations execution tasks should be invisible to the user including aliases task regexes and task requests from the command line

the correct authentication requirements arenoformatdbadduser user username pwd password roles otherdbroles local admin noformat

in a virtual environment with of data it uses of available ram just one connection to db and using the aggregation frameworkwhen i execute the query bellow the memory it is used without mesure and finally my db it is killed

hiwe got error in show dbs commandmongos show dbswed dec uncaught exception listdatabases failedassertion runcommand listdatabases on shard failed assertion cant map file memory assertioncode errmsg db assertion failure ok assertioncode db assertion failureok logs wed dec command admincmd command getlasterror dec runquery called admincmd listdatabases wed dec run command admincmd listdatabases wed dec command listdatabases wed dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec checking size file dec accessing for the first timewed dec couldnt open too many open fileswed dec user assertion map file memorywed dec command admincmd command listdatabases exception cant map file memory dec runquery called admincmd replsetgetstatus forshell wed dec run command admincmd replsetgetstatus forshell wed dec command replsetgetstatus forshell wed dec command admincmd command replsetgetstatus forshell dec insert dec runquery called admincmd getlasterror wed dec run command admincmd getlasterror wed dec command admincmd command getlasterror dec insert dec runquery called admincmd getlasterror wed dec run command admincmd getlasterror wed dec command admincmd command getlasterror dec socketexception remote error socket exception server wed dec runquery called admincmd whatsmyuri wed dec run command admincmd whatsmyuri wed dec command admincmd command whatsmyuri dec runquery called admincmd serverstatus wed dec run command admincmd serverstatus wed dec command admincmd command serverstatus dec socketexception remote error socket exception server wed dec runquery called admincmd whatsmyuri wed dec run command admincmd whatsmyuri wed dec command admincmd command whatsmyuri dec runquery called count database query fields wed dec run command count database query fields wed dec command command count database query fields dec socketexception remote error socket exception server wed dec runquery called admincmd getlasterror according above logs file descriptor limit has been exhaustedwe have runned following command system file descriptor limit sysctl fsfilemax fsfilemax shell files descriptor limit ulimit n current files descriptor are in used lsof wc l have following question what is the reason to come error too many file open is there any think wrongplease help meregardsjrv

hi there the createindex function in replicaset returns not master error you can find as attachments a simple snippet of js code to reproduce the issue in the example i used the secondary readpreference to make always return the error if it changes to primary it works and with the other preferences it has a random behaviour

paneltitleuseful author lead specification pocs ruby commit with required sdam changes commit with mandate that connection pools be cleared after not masterrecovery errors panel

this is caused by the fix for the ternary operator explicitly declares dbobject obj as null if no matching document is found but then with no test attempts to markaspartialobject which will always throw a nullpointerexception in the event findone found no matching document

code test jstestsreversecursorjs tue dec drop testjstestsreversecursor tue dec cmd drop testjstestsreversecursor tue dec build index testjstestsreversecursor id tue dec build index done scanned total records secs tue dec cmd validate testjstestsreversecursor tue dec validating index testjstestsreversecursoridtue dec exec error typeerror cannot read property of undefined return extentsfirstrecord null tue dec exec error error loading file jstestsreversecursorjs loadxname failed to load

this breaks drivers built on top mongoc there is also a todo in the code for this srcmongocmongoctopologyscannerc todo hook up a callback for this so you can use your own stream initiator

codemon oct uncaught exception assertsoon failed function var ok true try status catch e printe checkprimaryis reconnecting reconnectmaster status var str goal nodehost states if status statusmembers return false statusmembersforeachfunction m str mname mstate if mname nodehost ok mstate else ok mstate mstate mhealth print printstr print occasionallyfunction printnstatusprintjsonstatusprint return ok

introduced a new api for detecting conflicts among pos instances as part of that ticket to detect different migration id with same tenant id conflicts the tenant migration donortmd api was made to wait for the existing tmd instances initial state doc to be majority committed this can lead to deadlocks i think accidentally fixed that way deadlock issue but made the api primaryonlyservicecheckifconflictswithotherinstances racy racy scenario tmd instance migration id tenant id instance migration id tenant id calls getorcreateinstance acquires pos mutex calls checkifconflictswithotherinstances no conflicts detected create schedules instancerun to run asynchronously insert into pos inmemory map keyed using migrationid release pos mutex calls getorcreateinstance acquires pos mutex calls checkifconflictswithotherinstances no conflicts detected as instance durable state is empty create schedules instancerun to run asynchronously insert into pos inmemory map release pos mutex starts persists state doc with migration id tenant id durable state gets updated to nonempty starts persists state doc with migration id tenant id additional notes on deadlock scenario instance holds pos mutex as part of getorcreateinstance and wait for initial state doc to be majority committed stepdown thread holds rstl in mode x and tries to acquire pos mutex to execute pos onstepdown to interrupt active instances and blocks behind instance instance tries to acquire rstl in ix mode to write the initial state doc but blocks behind the stepdown thread

scott asked me to open this bug we have been seeing cursornotfound exceptions very infrequently in production i added some client side logging the result of which is the tolisttime is the time in ms that it took to read the entire cursor into memory we do this immediately after the query to get the closest thing to a snapshot view of the world when the query occurred we were getting odd data races otherwise

it is marked as collectible for legacy messages but not for modern updates with the pojocodec this could generate an id for the update part of the replace operation

the default mongoclient constructor tries to connect to instead of even if the constructor gets either mongoclientsettings or a connection string it does that strange thing of adding unspecified as a prefix to the host name

im trying to use insertbatch to process a file i use filereadlines to enumerate the lines in the file and turn them into bsondocuments i pass the linq enumerator to insertbatch it just hangs it works fine with a small file it does the same thing if i manually batch by using skip take on my linq query which is pulling lines out of the file if i take the first lines and pass them to insertbatch the first call works and the second batch just hangsi thought it might be something to do with one of these but am pretty stuck to be honest helpcode snippet var postcodestoinsert file readlinescsvfile select postcode new bsondocument postcode postcodesplitreplace stringempty replace stringempty postcodescollectioninsertbatchpostcodestoinsertthanksnic

params should be param field list should include roles after users

i installed mongocsharpdriver but on one operation in my project var queryeq var queryeq var queryin var queryeq var queryeq var var queryall var result collectionfindonequeryall i catched exception an exception of type systemnullreferenceexception occurred in mongodbbsondll but was not handled in user code additional information object reference not set to an instance of an object but if i changed this operation like this var result collectioncountqueryall worked well also this operation working properly on driver version and version what could it be

it is a common error for mongo newbies to attempt to run these commands from the shell it would be very useful to add a sentence or two to the beginning of the documentation explaining that these commands like mongod itself must be run from the system command line

trying to import the following csv codejava idreceivermonthaccrualmonthpaymentdatabankcodeoperation code when i run in mongo shell mongo version codejava mongoimport db meandev modemerge collection fulfilledpayments type csv headerline file downloadsimportcsv vvvv code it returns the following log but it doesnt really import codejava using upsert fields using decoding workers using insert workers will listen for sigterm sigint and sigkill filesize bytes using fields idreceivermonthaccrualmonthpaymentdatabankcodeoperation connected to localhost ns meandevfulfilledpayments connected to node type standalone standalone server setting write concern w to using write concern jfalse fsyncfalse standalone server setting write concern w to using write concern jfalse fsyncfalse got line imported document code but nothing is really imported into the database which remains untouched like this codejava id creator amounttransferred externalid date payments v code if i get the collection fulfilledpayments parameter off it imports to a new collection but of course there in no need for the merge mode there because it doesnt contain the id to be matched

presently the events for command monitoring include the exact commands sent over the wire in testing i deleted the clustertime and lsid fields to make viewing commands somewhat easier and it ended up changing internal state of the driver we should emit copies of these commands as part of the events

the db was configured as a capped collection from the beginning of to now the same kind of crash happened about times times this week and today crashed times cause its used as log db journal is disabled nojournalthe crash log looks like the following mon jul query trackentries jul query trackentries jul query trackentries jul insert trackentries jul insert trackentries jul query trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul invalid access at address mon jul got signal segmentation faultmon jul backtrace mon jul dbexit mon jul shutdown going to close listening socketsmon jul closing listening socket jul closing listening socket jul closing listening socket jul removing socket file jul got request after shutdownmon jul got request after shutdownmon jul shutdown going to flush diaglogmon jul shutdown going to close socketsmon jul got request after shutdownmon jul shutdown waiting for fs preallocatormon jul shutdown closing all filesmon jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul closeallfiles finishedmon jul shutdown removing fs lockmon jul dbexit really exiting nowlogstreamget called in uninitialized statemon jul error clientclient context should be null but is not clientconnthu jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul mem mb jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul insert trackentries jul invalid access at address thu jul got signal segmentation faultthu jul backtrace thu jul dbexit thu jul shutdown going to close listening socketsthu jul closing listening socket jul closing listening socket jul closing listening socket jul removing socket file jul shutdown going to flush diaglogthu jul shutdown going to close socketsthu jul shutdown waiting for fs preallocatorthu jul shutdown closing all filesthu jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul end connection jul closeallfiles finishedthu jul shutdown removing fs lockthu jul dbexit really exiting nowlogstreamget called in uninitialized statethu jul error clientclient context should be null but is not clientconn

regardingquery introspection improvementsmongodb includes a new query introspection system exposed in the explain method which is available on all mongodb query operations eg dbcollectionfind dbcollectionremove dbcollectionupdate dbcollectionfindandmodify operationsaccording to findandmodify is not yet covered by the new explain command

the documentation for text indexes mentions instead of textindexversion docs link

the attached code throws an invalidoperationexception the linqtomongoinject method is only intended to be used in linq where clauses var point new geojsonpointnew imongoquery query querygeointersectso ocoord point var result objectscollectionasqueryablewherex queryinjecttolist exception thrown here in version of c driver this code was worked i use linq where why exception was thrown

resolved current issue leaving open for better reporting when mongos auth fails to a sharded databasesee code mongo uusername ppasswordmongodb shell version to mar uncaught exception error err socket exception code exception login failedcodeseems related to and but they are on and it isnt working

sm api changed in between

some steps in restore the primary are referencing secondary servers see steps it appears that the primary server should be referenced there instead

we should extend the jstestfuzzer to support fuzz testing of internalunpackbucket in particular the agg fuzzer should be a good starting point

i cant reopen since was released so filing a new ticket

when performing a query against indexed subdocument fields we are able to retrieve and update the subdocument within an array using the operator there are cases where we would have need to know what subdocuments were located who satisfied our query and make use of that data a simple example would be to locate a document based on its subdocuments content and append a new subdocument that derives from that located subdocumentgiven subs id value id would like to execute something likedbcolupdate subsid push subs id value subsvalue that would result in appending a new subdocument in all documents who contain the matched subdocument from our query with a new value derived from the matched subdocument subs id value id value id please see for full discussionthis relates to but in that case the subdocuments are being updated this request is for the ability to access the subdocuments content who satisfied the query so we can make further decisionsmodifications based on that contentas an added advantage i could seen a lot of power added to javascripting if this were also available within js for examplefunction previousid newid dbcolfind subsid previousid foreach function thing previousvalue thingsubsvalue dbcolupdate id thingid push subs id newid value previousvalue in this case where the primary document was queried based on its subdocument contents it would be significant if we had access to that data through the passed in matched document i am not sure if this would be the ideal syntax but having some way to reference the specific subdocument that satisfied the this outer query can give a lot of power to the inner for making decisions based on such content

mobile se with concurrent jstests asserts saying database is locked needs further digging into and fixing as needed also several tests on mobile se variant are marked to be skipped due to this bug they need to be unskipped

currently we do not allow users to insert nongeo data for example a string into a field that has a geo index on it here is an examplenoformat dbtestensureindexa createdcollectionautomatically false numindexesbefore numindexesafter ok dbtestremovewriteresult nremoved dbtestinsertid a this is not a testwriteresult ninserted writeerror code errmsg insertdocument caused by cant extract geo keys id a this is not a test geo element must be an array or object a this is not a test noformati could not find this limitation documented anywhere we should make this clear to the users as it is a fairly major limitations of geo indexes

jstestsshardingtxnreadonlytransactionsjs tests read only transactions by running committransaction commands and asserting certain lines are logged by mongos to determine the correct version of commit was used ie a readonly commit vs a twophase commit it does so by searching for the expected log in the shells rawmongoprogramoutput after running the commits although the commits must have finished by this point it is not guaranteed the lines logged by mongos during commit have reached the shells rawmongoprogramoutput in which case the test fails verified by adding sleeps in programoutputmultiplexerappendline when the line to add is one of those the test is looking for the simplest fix for this is to perform the search for logs in a loop since in normal operation the lines should reach rawmongoprogramoutput eventually

we met the bug after we upgraded our mongos to the mongos still crashes code i asio connecting to i invariant failure optimedout srcmongoexecutornetworkinterfaceasiocpp i aborting after invariant failure i asio failed to connect to exceededtimelimit operation timed out f got signal aborted begin backtrace backtraceprocessinfo mongodbversion gitversion compiledmodules uname sysname linux release version smp wed oct utc machine somap mongos mongos mongos mongos mongos end backtrace i asio connecting to i asio connecting to i asio connecting to i asio connecting to i asio connecting to i asio connecting to i asio connecting to code

we should have tests that create option structs with all supported options and assert the encoded result this would have allowed us to catch earlier

when a backup daemon fails to resync the onpremise documentation does not document this event need to document how this is detected steps we take to recover and why its a bad idea to automatically recover from this event

is there any way to reduce the default command timeout of of runcommand in below code codejava var command new bsondocument dbstats var stats databaseruncommandcommand consolewritelinestatstojson do something with the stats code thanks in advance irshad

if you think this change is important enough to include a snippet or link on the driver landing pages eg somewhere after connect to mongodb and compatibility on a page like please file a docsp ticket and link it to this ticket document how to create a client that declares an api version include example document the strict option functionality include example document the deprecation errors option include example document generic command helper behavior include example heres pseudocode andreas created that can be used for other drivers declare an api version on a client code declare api version for the client serverapi new client new mongoclienturi serverapi cursor clientdatabasecollectionfind code strict option declaring a strict api version will cause the mongodb server to reject all commands that are not part of the declared api version this includes command options and aggregation pipeline stages for example the following find call would fail because the tailable option is not part of version code declare api version for the client serverapi new strict true client new mongoclienturi serverapi fails with an error cursor clientdatabasecollectionfind tailable true code the deprecationerrors option can be used to enable command failures when using functionality that is deprecated from version note that at the time of this writing no deprecations in version exist code declare api version for the client serverapi new deprecationerrors true client new mongoclienturi serverapi code the declared api version is applied to all commands run through the client including those sent through the generic command helper specifying versioned api options in the command document and declaring an api version on the client is not supported and will lead to undefined behaviour to run any command with a different api version or without declaring one create a separate client that declares the appropriate api version

audit the driver for places that should say noexcept that currently dont and apply where neededold descriptiona few methods in the legacy driver have dynamic exception specifiers such asoperationexceptionoperationexceptionoperationexceptionwhatsspiclientmechnewsspiclientmechstopdbexceptiondbexceptiondbexceptionwhatassertionexceptionassertionexceptionsocketexceptionsocketexceptionwe should remove them and consider replacing with a macro that is defined to be the noexcept operator on compilers that support it and a noop otherwise

paneltitledownstream change scope document implement snapshot session spec implement snapshot session tests this ticket introduced unified tests for the sessions specification drivers should sync tests in sessionstestsunified with existing legacy tests were moved to sessionstestslegacy but are otherwise unchanged panel description of linked ticket paneltitleepic summary support readconcern level snapshot nonspeculative for read commands outside of transactions including on secondaries motivation once durable history is implemented in wiredtiger mongodb will be able to support longrunning snapshot reads by supporting snapshot reads on secondaries this feature can be used for analytics queries its important to use nonspeculative behavior so that queries do not need to consume all results and wait for a transaction to commit in order to know that the data read was majority committed documentation scope document technical design document panel html spec ticket description this block of html and the script in it autopopulate the spec ticket description on page load script target if you can read this text the script has failed get functiondata var description datafinddescriptionval descriptionfindscripttargetwrapperremove scripttargethtmldescription html

bsonerrorsinvaliddocument key date must not start with

i was running a culster named clustergreen with servers main servers and one arbiteri deleted all the databases rm on disk on the four main servers but left the server the arbiter unchangedi then configured the four main servers to use a different replset called greenwhen i restarted the arbiter configured to replset green the server crashedsee attached log file

the slave creates the default id index spec for the collection here instead it should copy the exact id index spec from master

tldr with some performance tests we can see if we can deprecate opquery and opgetmore in favor of find and getmore opquery and opgetmore are legacy for fcv servers and drivers charlie swanson wrote in i believe there was some noticeable regression of the find and getmore commands compared to opquery and opgetmore i asked for more and he wrote the summary is that the query team would love to deprecate the legacy wire protocol and move on to find and getmore commands for everything our understanding is that the command versions are the default in modern drivers but no one is comfortable deprecating the old ones because there was some performance regression when first testing find and getmore command vs the old opquery and opgetmore i have heard rumors that further investigation proved the server wasnt as much to blame as we thought it was but we havent followed up to build new performance tests and validate that performance is comparable supporting opquery and opgetmore has slowed development of many query projects the most recent example i can think of is change streams we currently have two implementations for iterating a cursor opgetmore and getmore command are completely independent and definitely duplicate a lot of logic we have postponed investing in refactoring this in hopes of deprecating opgetmore see

version released the default value for sslrequirevalidcertificates has been changed to true users upgrading from and using ssl will need to set the value of ssltrustedservercertificates in their configuration file more information can be found here in the docs

on this page think all we need to say is that people need to install the new repo files basically following the installation instructions as if they dont already have packages installed the old repos shouldnt conflict with new ones and with both oldnew installed the newer packages in the new repos should just upgrade over the old ones as if they were from the same repo cc

in the interest of usability we should use type maps to ensure that embedded documents are converted to an object implementing arrayaccess eg arrayobject instead of stdclass objects this would make a fine default for users although we should make it easy for them to specify a typemap at the client level and have it be inherited for database and collection operations

quote i cant speak for other platforms but with my current setup trying to do will save an in the database with value so it wraps around instead of truncating quote quote inc suffers from the same problem as set that is if i try to i will actually subtract from testint further testing has revealed that queries suffer from the same problem if i were to make a query with a filter like testint ill actually be looking for documents where testint is less than quote truncation of integers when converting bson to php values may also result in property name corruption see

we used an application named monstache with mongogodriver to watch change streams from replica ret with mongods of and the whole system works well for a few weeks but the primary mongod restarted frequently after we started to use transaction with javagodriver we tried the following conditions to check the cause when the server restarted frequently we check the connection status between hostpost and not this we checked and ensured that the wach apis of mongogodriver are used correctly the mongo processes are started with systemd and we add the remommended ulimits in the systemd service and we created mongoclient instances and ran our transaction unit tests with monstache running but the mongo didnt restart however sometimes the mongod restarted all of a sudden below are the logs and the key words are server restarted and the ctxconnxxx which is generated by mongogodriver used by monstache and the printed backtrace and invariant failure

an improvement from was to enable the scons j n option to get multithreaded compiles a side effect is that scons check is also multithreaded and that can cause a collision when the same temp directory is used by two programs in this failure exprocess reports a problem ive seen it in a pr run where exthread reports the same problem sconscheck failed on windows host project wiredtiger develop commit diff use the correct python for all tests and set up env for the split stress test oct utc evergreen subscription evergreen event task logs

in the access control sectionif you are using mongodb authentication see the authentication requirements documentation

of string created like this in pythonin dbfoodroptrue b new dbfooinsertbb nindexeswas ok problem is in constructor we need to check if len before this line

we need to come up with a way to create client error codes that do not overlap with server error codes in the futureexisting error codes can probably be shared

description in our docs page for the mongo client we have a couple of examples for connecting with various options it would be a good idea to add some more examples for more complex connection strings for example kerberos authentication noformatmongo noformat special character in the password noformatmongo noformat the same location in the docs also contains double quotes around all uris the examples all work across platforms but this can cause problems if there are nonencoded special characters in the uri for nix and windows powershell you should use single quotes and double quotes for the windows cmd prompt it might be a good idea to add this clarification also scope of changes impact to other docs mvp work and date resources scope or design docs invision etc

code to reproduce codejavascript const mongoclient requiremongodbmongoclient const inner a b mongoclientconnect usenewurlparser true thenclient clientdbdbcollectioncol thencollection collectioninsertoneinner thenconsolelog catchconsoleerror then processexit code expected document inserted as specified in field limitations official documentation bq field names can contain dots ie and dollar signs ie however only nested field names may start with a dollar sign for example attempting to insert a b will succeed but attempting to insert a results in error actual throws error noformat error key b must not start with at serializeinto at serializeobject at serializeinto at serializeobject at serializeinto at serializeobject at serializeinto at bsonserialize at querytobin at serializecommands noformat

code fri feb databaseholdercloseall pathdatadbcloseall fri feb info journal rotate called inside dbmutex ok but should be somewhat rare fri feb durparanoid mismatch in

in the current mapreduce implementation there is nothing in place to explicitly reject merging to an existing sharded collection whose shard key is not id however since it uses this upsert helper which always targets by id the command will eventually fail we should keep this behavior in

there is a panic that is caused in the changestreamsexamplestest it happens infrequently and on random evergreen patches

i realize there is a note to replace localhost with a real routable address but people keep missing it and they cutnpaste stuffrequest is to replace localhost with some string that wont work as is

original title empty match at the start of the aggregation pipeline greatly impacts performance in does not in aggregation below seconds seconds removing the empty match seconds seconds code dbcollectionaggregate match project docfield id unwind doc unwind unwind group id countid addtoset id countvariationsvariationid sum code

dbmycolldrop cannot release table spaces and when i restart the mongod service the space is released

when i rename a field in mongoid it successfully updates the field in mongodb but it does not update the field in the ruby code class example include mongoiddocument field apple type string as capple end exampleall mongoidcriteria selector options class example embedded false exampleallrenameapple dapple moped update databasecore collectionexamples selector updaterenameappledapple flags command databasecore runtime updatedexistingfalse writtentonil errnil e examplenew edapple nomethoderror undefined method dapple for

this is a blocker weve stumbled upon while upgrading to bson the following script demonstrates code require bson class hi hash end a puts afirstclass hi atobsonnormalizedvalue puts afirstclass bsondocument code ive opened a pr that fixes the issue thanks

mongo c driver does not return consistent results with starts with regex query i made a query to my database through mongo studio however my c client does not get the same results as mongo studio sometimes it retrieves some results other times it does not return any results the image displays a scenario where no results are returned the query is id defaultcampaignssystemi

begin backtrace backtraceprocessinfo mongodbversion gitversion compiledmodules uname sysname linux release version smp thu jul utc machine somap mongod mongod end backtrace

including ssl false attempts ssl connection the only way to avoid an ssl connection is no ssl option at all

we have around app servers on which we are running mongos which connect to config serversthere was following error in one of the mongos server caught exception while doing balance error checking clock skew of cluster caused by clock skew of the cluster is too far out of bounds to allow distributed lockingthis is due to time difference but we have ntpd service running the difference between time of one working mongos server to nonworking mongos is around which i dont think should create this sep distributed lock unlocked mon sep distributed lock acquired ts sep distributed lock unlocked this is coming on one of the mongos server wanted to confirm is balancing works only on one of the mongos serveralso dblocksfind id balancer pretty gave following output id balancer process state ts when who why doing balance roundso to summaries all mongosa b c show different status in a no log is generatedworking fine in b clock skew issue is comingnot working correctly in c distributed lock issue is coming

a number of json driven tests has a bug when they always fail in async tests for examplethere are more of them jsondrivenassertsessionpinnedtest the only logic in the test is asserting however since actasync is not overwritten calling this test from the async mode will always trigger notimplemented exception the only reason why we dont see this exception because we overwrite the base implementation for assert and as a result we just ignore the thrown exception

when saving an expandoobject that contains a guid property guidserializer throws an exception if bsondefaultsguidrepresentationmode see this code mongocsharpdriverguidserializercs at master · mongodbmongocsharpdriver githubcom exception is thrown at line codejava throw new bsonserializationexceptionguidserializer cannot serialize a guid when guidrepresentation is unspecifiedcode the following code reproduces the issue codejava bsondefaultsguidrepresentationmode var client new var db clientgetdatabasetest var coll dbgetcollectionexpando var exp new expandoobject idictionaryexp guidnewguid await collinsertoneasyncexpcode if the first line is omitted the expandoobject is serialized and saved to db but not as a uuid but as binary eg codejava code

there are a couple of shell scripts with odd syntax after the migration see the following links for detail missing space after extra

i just wanted to point out a new driver for delphi optimized with mongodb it can store documents in variants with latebinding for the properties and handle all bson types it has also full json support with mongodb extended syntax and has been optimized for speed from the ground upsee there is also an orm layer available and a clientserver restul access via json resulting benchmark is awesome when compared to regular sql engines monodb rocks

add iamgetuser to minimum access policyiam block of access policy should look something like the belownoformat effect allow action resource noformat

this started to happen after pulling the current main not sure but this should be the commit before i had already unlocked the entries in mongodb compass devconnections we should verify that this issue would not affect also beta and ga packages before proceeding with a release

hi mongoexport documents has been deleted form mongodborg please submit them on site the uri is

there are a few updates needed for onprem

migrationmanagerabandonactivemigrationsandenablemanager should also accept kstopping and put it in the appropriate state it shouldnt abandon the recovery clear configmigrations and release distlocks in this case just halt the recovery

isolates just means that once a multiupdate has written it can no longer yield i started to clean things up to make that clear but more work may be needed commits so farthis will change significantly in so these edits are most important for

add support for google analytics or similar usage statistics platform to enable collection of information for how users interact with the various evergreen pages

the installation procedure contains this command line to configure selinuxsemanage port a t mongodbportt p tcp the port type should be mongodportt without the b

we should extend tests that live in jstestsmultiversiongenericsetfcvusage to use lastlts and lastcontinuous we should remove any tests that we are confident will be covered under our multiversion passthrough testing

i submitted two patches today and while compile completed they have been stuck in this state for hours submitted at noon just cancelled submitted at