cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jonathan Ellis (JIRA)" <j...@apache.org>
Subject [jira] Commented: (CASSANDRA-1991) CFS.maybeSwitchMemtable() calls CommitLog.instance.getContext(), which may block, under flusher lock write lock
Date Mon, 24 Jan 2011 14:55:46 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-1991?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12985717#action_12985717
] 

Jonathan Ellis commented on CASSANDRA-1991:
-------------------------------------------

Taking a step back, this feels like a _lot_ of extra complexity added to one of our already-complex
subsystems, to fix a fairly minor problem (in the sense that it can only occur on flush, which
are supposed to be tuned to be rare).

How can we reduce the fragility (in the sense that it's easy to do the wrong thing when editing
code)?

> CFS.maybeSwitchMemtable() calls CommitLog.instance.getContext(), which may block, under
flusher lock write lock
> ---------------------------------------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-1991
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-1991
>             Project: Cassandra
>          Issue Type: Improvement
>            Reporter: Peter Schuller
>            Assignee: Peter Schuller
>         Attachments: 1991-checkpointing-flush.txt, 1991-logchanges.txt, 1991-trunk-v2.txt,
1991-trunk.txt, 1991-v3.txt, 1991-v4.txt, 1991-v5.txt, 1991-v6.txt, 1991-v7.txt, 1991-v8.txt,
1991-v9.txt, trigger.py
>
>
> While investigate CASSANDRA-1955 I realized I was seeing very poor latencies for reasons
that had nothing to do with flush_writers, even when using periodic commit log mode (and flush
writers set ridiculously high, 500).
> It turns out writes blocked were slow because Table.apply() was spending lots of time
(I can easily trigger seconds on moderate work-load) trying to acquire a flusher lock read
lock ("flush lock millis" log printout in the logging patch I'll attach).
> That in turns is caused by CFS.maybeSwitchMemtable() which acquires the flusher lock
write lock.
> Bisecting further revealed that the offending line of code that blocked was:
>                     final CommitLogSegment.CommitLogContext ctx = writeCommitLog ? CommitLog.instance.getContext()
: null;
> Indeed, CommitLog.getContext() simply returns currentSegment().getContext(), but does
so by submitting a callable on the service executor. So independently of flush writers, this
can block all (global, for all cf:s) writes very easily, and does.
> I'll attach a file that is an independent Python script that triggers it on my macos
laptop (with an intel SSD, which is why I was particularly surprised) (it assumes CPython,
out-of-the-box-or-almost Cassandra on localhost that isn't in a cluster, and it will drop/recreate
a keyspace called '1955').
> I'm also attaching, just FYI, the patch with log entries that I used while tracking it
down.
> Finally, I'll attach a patch with a suggested solution of keeping track of the latest
commit log with an AtomicReference (as an alternative to synchronizing all access to segments).
With that patch applied, latencies are not affected by my trigger case like they were before.
There are some sub-optimal > 100 ms cases on my test machine, but for other reasons. I'm
no longer able to trigger the extremes.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message