Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 12:40:47 +0200] rev 4014
sqlcache: also catch malformed database error
This is apparently another way for sqlite to fail at concurrency.
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 12:40:41 +0200] rev 4013
stablerange: rework saving logic to be clearer and safer
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 12:12:49 +0200] rev 4012
obshashrange: rework saving branching to be clearer and safer
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 11:40:32 +0200] rev 4011
sqlcache: protect read query too
Some error (like locked database) can even happens when doing readonly
operation. So we protect them too.
At that point, it seems that pysqlite3 is not the right tool for this job.
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 11:33:09 +0200] rev 4010
sqlite: fast path when nothing to save
If we have no new data to save, we should try to do anything. Doing something
concurrently is hard enough. If we can avoid doing it at all that would be
great.
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 11:27:04 +0200] rev 4009
sqlcache: initialize meta table last
It turns out that pysqlite commit after each table creation. This commit is an
hopeless attemps to make the caches more concurrent friendly.
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 10:20:15 +0200] rev 4008
obshashrange: always check in base cachekey against the recorded one
Ignoring the check in the none case is just disaster waiting to happens.
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 10:16:12 +0200] rev 4007
stablerangecache: fix output in the drifted case
The node were not hex, and the field had different order in the pair. This is
now fixed.
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 00:35:51 +0200] rev 4006
sqlcache: also ignore integrity error
integrity error can happens when multiple client tries to warm similar cache.
Given we leave multiple hole in the cache that can be warmed by anyone, this is
harder to avoid that we would like. We simply catch the error in this case.
Someone will warm the missing entry later.
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 00:28:19 +0200] rev 4005
sqlcache: cache OperationError when saving
This is a cache, so we should not crash when trying to operate on it.
OperationError can happens when the database is seen as readonly or locked.
Pierre-Yves David <pierre-yves.david@octobus.net> [Mon, 27 Aug 2018 00:18:06 +0200] rev 4004
sqlcache: passe better connection option
These new options should help with handling transaction consistency and database
access on high load.
Pierre-Yves David <pierre-yves.david@octobus.net> [Sun, 26 Aug 2018 20:55:26 +0200] rev 4003
contrib: introduce a small script to stress tests obsolescence exchange
Tester have been reporting some error that seems to originate from concurrent
access/update to the cache. This script will be useful to reproduce these
situations locally.
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 21 Aug 2018 14:09:33 +0200] rev 4002
test-compat: merge mercurial-4.4 into mercurial-4.3
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 21 Aug 2018 13:59:27 +0200] rev 4001
test: adjust glob in previous changeset
I forgot to glob the time.
Pierre-Yves David <pierre-yves.david@octobus.net> [Tue, 21 Aug 2018 15:32:02 +0200] rev 4000
branching: merge with stable