-
Notifications
You must be signed in to change notification settings - Fork 2.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Messaging: Avoid deadlocks related to 0 receiver behavior #10132
Conversation
Signed-off-by: Matt Lord <[email protected]>
d4ff217
to
f31eefa
Compare
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
f680779
to
d177a21
Compare
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
See: vitessio/website#1015 Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Looks like the majority of tests are failing on
|
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
This is for safe concurrency with the last receiver unsubscribing Signed-off-by: Matt Lord <[email protected]>
@mattlord you really rolled through quite a few iterations, thanks again for digging in so deep. I don't think I have enough context around vstream to provide much feedback on the latest design. Once @rohit-nayak-ps approves it, I'm ready to roll it out in prod to test it out. |
Signed-off-by: Matt Lord <[email protected]>
Signed-off-by: Matt Lord <[email protected]>
Note that the |
@mattlord is this still waiting on @rohit-nayak-ps, or was pairing with @sougou sufficient review? |
🎉 Thanks so much for tracking this down!! I'll be rolling it out later today |
Excellent! Please let me know how things go. |
This reverts commit 5e02d6e.
* Prevent deadlocks related to 0 receiver behavior Signed-off-by: Matt Lord <[email protected]> * Update test tables to use poller_idx Signed-off-by: Matt Lord <[email protected]> * Minor changes after mutex usage review in message manager + cache Signed-off-by: Matt Lord <[email protected]> * Use atomics for receiver count and messages pending Signed-off-by: Matt Lord <[email protected]> * Don't take exclusive lock when in fast path Signed-off-by: Matt Lord <[email protected]> * Update tests to use the new recommended message table structure See: vitessio/website#1015 Signed-off-by: Matt Lord <[email protected]> * Correct tests Signed-off-by: Matt Lord <[email protected]> * Update e2e test to use new recommended table structure Signed-off-by: Matt Lord <[email protected]> * Fix TestMessageStreamingPlan test Signed-off-by: Matt Lord <[email protected]> * Fix godriver/TestStreamMessaging test Signed-off-by: Matt Lord <[email protected]> * Split streamMu into streamProcessingMu and lastPollPositionMu Signed-off-by: Matt Lord <[email protected]> * Poller cannot take main lock w/o having X stream processing lock Signed-off-by: Matt Lord <[email protected]> * Improve the comments a bit Signed-off-by: Matt Lord <[email protected]> * Hold the main mutex during Add This is for safe concurrency with the last receiver unsubscribing Signed-off-by: Matt Lord <[email protected]> * Changes after pair reviewing with Sugu Signed-off-by: Matt Lord <[email protected]> * Use my GitHub handle for the self reference Signed-off-by: Matt Lord <[email protected]> Signed-off-by: Rameez Sajwani <[email protected]>
…11619) * Move towards MySQL 8.0 as the default template generation (#11153) * Move towards MySQL 8.0 as the default template generation This upgrades the remaining things to Ubuntu 20.04 and makes MySQL 8.0 the default we run tests against. We still have tests for MySQL 5.7 but those are now explicitly opted into. This should finish up the Ubuntu 20.04 upgrade and also makes things easier for the future when we need to upgrade again. Signed-off-by: Dirkjan Bussink <[email protected]> * CI: rename shard vtorc_8.0 to vtorc_5.7, change expected test output for 8.0 Signed-off-by: deepthi <[email protected]> * CI: increase timeout for 8.0 tests on the actual test step from 30 to 45 mins Signed-off-by: deepthi <[email protected]> * CI: increase timeout to 45 minutes for mysql57 tests too. We really only need this for vtorc, but I've made the change to the template so all tests get it. Signed-off-by: deepthi <[email protected]> * CI: fix vtorc test to work with both 5.7 and 8.0 Signed-off-by: deepthi <[email protected]> * CI: missed docker flag in mysql57 template, one more fix to vtorc test Signed-off-by: deepthi <[email protected]> * removing spaces from pb file Signed-off-by: Rameez Sajwani <[email protected]> * removing spaces in pb file part 2 Signed-off-by: Rameez Sajwani <[email protected]> Signed-off-by: Dirkjan Bussink <[email protected]> Signed-off-by: deepthi <[email protected]> Signed-off-by: Rameez Sajwani <[email protected]> Co-authored-by: deepthi <[email protected]> Co-authored-by: Rameez Sajwani <[email protected]> * fixing template files to default to mysql80 Signed-off-by: Rameez Sajwani <[email protected]> * changing update statement to alter Signed-off-by: Rameez Sajwani <[email protected]> * Avoid deadlocks related to 0 receiver behavior (#10132) * Prevent deadlocks related to 0 receiver behavior Signed-off-by: Matt Lord <[email protected]> * Update test tables to use poller_idx Signed-off-by: Matt Lord <[email protected]> * Minor changes after mutex usage review in message manager + cache Signed-off-by: Matt Lord <[email protected]> * Use atomics for receiver count and messages pending Signed-off-by: Matt Lord <[email protected]> * Don't take exclusive lock when in fast path Signed-off-by: Matt Lord <[email protected]> * Update tests to use the new recommended message table structure See: vitessio/website#1015 Signed-off-by: Matt Lord <[email protected]> * Correct tests Signed-off-by: Matt Lord <[email protected]> * Update e2e test to use new recommended table structure Signed-off-by: Matt Lord <[email protected]> * Fix TestMessageStreamingPlan test Signed-off-by: Matt Lord <[email protected]> * Fix godriver/TestStreamMessaging test Signed-off-by: Matt Lord <[email protected]> * Split streamMu into streamProcessingMu and lastPollPositionMu Signed-off-by: Matt Lord <[email protected]> * Poller cannot take main lock w/o having X stream processing lock Signed-off-by: Matt Lord <[email protected]> * Improve the comments a bit Signed-off-by: Matt Lord <[email protected]> * Hold the main mutex during Add This is for safe concurrency with the last receiver unsubscribing Signed-off-by: Matt Lord <[email protected]> * Changes after pair reviewing with Sugu Signed-off-by: Matt Lord <[email protected]> * Use my GitHub handle for the self reference Signed-off-by: Matt Lord <[email protected]> Signed-off-by: Rameez Sajwani <[email protected]> * remove unwanted sleep Signed-off-by: Rameez Sajwani <[email protected]> * Fix failures Signed-off-by: Rameez Sajwani <[email protected]> * Fix backup tests Signed-off-by: Rameez Sajwani <[email protected]> * fix vtgate_gen4 error Signed-off-by: Rameez Sajwani <[email protected]> * upgrading to mysql 8.0 using vinalla mysql Signed-off-by: Rameez Sajwani <[email protected]> * pin to specific version 8.0.25 Signed-off-by: Rameez Sajwani <[email protected]> * Try to pin mysql version to 8.0.25 using tar file Signed-off-by: Rameez Sajwani <[email protected]> * Fix bug in last commit for 8.0.25 Signed-off-by: Rameez Sajwani <[email protected]> * fixing template files to use mysql8.0.25 Signed-off-by: Rameez Sajwani <[email protected]> * fixing community version Signed-off-by: Rameez Sajwani <[email protected]> * removing all mysql version before installation Signed-off-by: Rameez Sajwani <[email protected]> * moving cluster 11 to self host Signed-off-by: Rameez Sajwani <[email protected]> * trying different combination since mysql is not getting installed correctly Signed-off-by: Rameez Sajwani <[email protected]> * use tar instead of deb file Signed-off-by: Rameez Sajwani <[email protected]> * remove mysql stop statement Signed-off-by: Rameez Sajwani <[email protected]> * setting vt_mysql_root Signed-off-by: Rameez Sajwani <[email protected]> * moving export to right place Signed-off-by: Rameez Sajwani <[email protected]> * change template to accomodate tar file download Signed-off-by: Rameez Sajwani <[email protected]> * fixing mysql80 cluster Signed-off-by: Rameez Sajwani <[email protected]> * Adding mysql to the path env variable Signed-off-by: Rameez Sajwani <[email protected]> * fixing shardedpitr_tls test Signed-off-by: Rameez Sajwani <[email protected]> * fix upgrade-downgrade test Signed-off-by: Rameez Sajwani <[email protected]> * adjust epected AUTO_INCREMENT value Signed-off-by: Shlomi Noach <[email protected]> * removed 'expected_table_structure' files because there are different outputs in mysql57 and in mysql80 Signed-off-by: Shlomi Noach <[email protected]> * adding mysql version in workflow logs Signed-off-by: Rameez Sajwani <[email protected]> * move to utuntu 18 for upgrade-downgrade test Signed-off-by: Rameez Sajwani <[email protected]> Signed-off-by: Dirkjan Bussink <[email protected]> Signed-off-by: deepthi <[email protected]> Signed-off-by: Rameez Sajwani <[email protected]> Signed-off-by: Matt Lord <[email protected]> Signed-off-by: Shlomi Noach <[email protected]> Co-authored-by: Dirkjan Bussink <[email protected]> Co-authored-by: deepthi <[email protected]> Co-authored-by: Matt Lord <[email protected]> Co-authored-by: Shlomi Noach <[email protected]>
Description
There's conditional behavior and code paths in the message manager when the number of receivers is 0. The main mutex and the stream mutex locks are used to ensure overall consistency. But:
vitess/go/vt/vttablet/tabletserver/messager/message_manager.go
Lines 752 to 787 in c985379
These two factors can lead to deadlocks when the receiver count goes to/from 0 while the the poller runs:
In
unsubscribe()
we have the main mutex, and while holding it we then callstopVStream()
if the subscriber/receiver count is at 0. ThenstopVStream()
tries to get the stream mutex.Concurrently,
runPoller()
— which by default, or more accurately when using the documented message table structure here via thevt_poller_interval
comment directive, runs on a 30 second ticker — gets the stream mutex first, then it later tries to get the main mutex.Now we are deadlocked. They're both waiting on each other.
This then in turn causes things like
PlannedReparentShard
to hang because during the transition from primary serving we stop/close the message service, which in turn will wait trying to get the main mutex.Changes
cacheManagementMu
lock. The explicit usage of this new mutex allows us to avoid the deadlocks seen before because it's taken immediately when the poller or binlog streamer process data coming from the database (which is used to update the cache), and we no longer need to take this lock when starting or stopping the binlog streamer itself.processRowEvent()
->Add()
— we were at risk of a deadlock. The main mutex became VERY hot because we were checking the receiver count all the time to ensure the fast path (do nothing when there's no receivers).time_acked is null
— to try and limit the size of scans done on the underlying message table to try and ensure that the poller runs quickly. This aspect is important because the binlog event processing is blocked during this time. It also prevents us from doing unnecessary work as InnoDB does most of the work — all query (WHERE) predicates become index condition pushdowns and the newpoller_idx
index can be used for the ordering to apply theLIMIT
as well — and passes the minimum amount of potentially useful data up the stack InnoDB->MySQL->MessageManager->Cache.Notes
This is able to complete on
vitessio/main
:Whatever changes are made here we should continue to pass that test (it does). It's even more reliable now — you can run it 1,000 times and no flakiness.
Related Issue(s)
Checklist