Tarantool development patches archive
 help / color / mirror / Atom feed
From: Sergey Petrenko via Tarantool-patches <tarantool-patches@dev.tarantool.org>
To: Vladislav Shpilevoy <v.shpilevoy@tarantool.org>, gorcunov@gmail.com
Cc: tarantool-patches@dev.tarantool.org
Subject: Re: [Tarantool-patches] [PATCH v4 12/16] box: fix an assertion failure after a spurious wakeup in promote
Date: Fri, 23 Jul 2021 09:45:22 +0200	[thread overview]
Message-ID: <4f192cbf-5bad-4eb7-9ed0-2789d5d4c7d8@tarantool.org> (raw)
In-Reply-To: <a08b0288-33f2-122a-6389-26b3bb3c8921@tarantool.org>


22.07.2021 01:29, Vladislav Shpilevoy пишет:
> I appreciate the work you did here!
>
> See 2 comments below.
>
>> diff --git a/test/replication/gh-3055-promote-wakeup-crash.result b/test/replication/gh-3055-promote-wakeup-crash.result
>> new file mode 100644
>> index 000000000..e508611e5
>> --- /dev/null
>> +++ b/test/replication/gh-3055-promote-wakeup-crash.result
>> @@ -0,0 +1,43 @@
>> +-- test-run result file version 2
>> +test_run = require('test_run').new()
>> + | ---
>> + | ...
>> +--
>> +-- gh-3055 follow-up: box.ctl.promote() could crash on an assertion after a
>> +-- spurious wakeup.
>> +--
>> +_ = box.space._cluster:insert{2, require('uuid').str()}
>> + | ---
>> + | ...
>> +box.cfg{election_mode='manual',\
>> +        replication_synchro_quorum=2,\
>> +        election_timeout=1000}
> 1. Shouldn't you save and restore the old option values in the end
> of the test?

Not anymore. As far as I know the default instance is restarted

between each test run now.

So we only need to reset persistent things now, like

grants, roles, promotes/demotes, spaces and so on.

>> + | ---
>> + | ...
>> +
>> +fiber = require('fiber')
>> + | ---
>> + | ...
>> +f = fiber.create(function() box.ctl.promote() end)
>> + | ---
>> + | ...
>> +f:set_joinable(true)
> 2. But you never use :join(). Perhaps you might want to call it
> after :cancel(). To ensure the fiber really ended.

Yep, my bad. Thanks for noticing!

=====================================

diff --git a/test/replication/gh-3055-promote-wakeup-crash.result 
b/test/replication/gh-3055-promote-wakeup-crash.result
index e508611e5..f915bf359 100644
--- a/test/replication/gh-3055-promote-wakeup-crash.result
+++ b/test/replication/gh-3055-promote-wakeup-crash.result
@@ -35,6 +35,11 @@ fiber.yield()
  f:cancel()
   | ---
   | ...
+f:join()
+ | ---
+ | - false
+ | - fiber is cancelled
+ | ...
  box.cfg{election_mode='off'}
   | ---
   | ...
diff --git a/test/replication/gh-3055-promote-wakeup-crash.test.lua 
b/test/replication/gh-3055-promote-wakeup-crash.test.lua
index 2ac901b08..acf0c05b0 100644
--- a/test/replication/gh-3055-promote-wakeup-crash.test.lua
+++ b/test/replication/gh-3055-promote-wakeup-crash.test.lua
@@ -16,5 +16,6 @@ fiber.yield()

  -- Cleanup.
  f:cancel()
+f:join()
  box.cfg{election_mode='off'}
  test_run:cleanup_cluster()

=====================================

>> + | ---
>> + | ...
>> +f:wakeup()
>> + | ---
>> + | ...
>> +fiber.yield()
>> + | ---
>> + | ...
>> +
>> +-- Cleanup.
>> +f:cancel()
>> + | ---
>> + | ...
>> +box.cfg{election_mode='off'}
>> + | ---
>> + | ...
>> +test_run:cleanup_cluster()
>> + | ---
>> + | ...

  reply	other threads:[~2021-07-23  7:47 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-14 18:25 [Tarantool-patches] [PATCH v4 00/16] forbid implicit limbo ownership transition Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 01/16] replication: always send raft state to subscribers Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 02/16] txn_limbo: fix promote term filtering Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 03/16] txn_limbo: persist the latest effective promote in snapshot Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 04/16] replication: encode version in JOIN request Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 05/16] replication: add META stage to JOIN Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 06/16] replication: send latest effective promote in initial join Serge Petrenko via Tarantool-patches
2021-07-21 23:24   ` Vladislav Shpilevoy via Tarantool-patches
2021-07-23  7:44     ` Sergey Petrenko via Tarantool-patches
2021-07-26 23:43       ` Vladislav Shpilevoy via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 07/16] replication: send current Raft term in join response Serge Petrenko via Tarantool-patches
2021-07-21 23:24   ` Vladislav Shpilevoy via Tarantool-patches
2021-07-23  7:44     ` Sergey Petrenko via Tarantool-patches
2021-07-26 23:43       ` Vladislav Shpilevoy via Tarantool-patches
2021-07-29 20:46         ` Sergey Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 08/16] raft: refactor raft_new_term() Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 09/16] box: split promote() into reasonable parts Serge Petrenko via Tarantool-patches
2021-07-21 23:26   ` Vladislav Shpilevoy via Tarantool-patches
2021-07-23  7:45     ` Sergey Petrenko via Tarantool-patches
2021-07-26 23:44       ` Vladislav Shpilevoy via Tarantool-patches
2021-07-29 20:46         ` Sergey Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 10/16] box: make promote always bump the term Serge Petrenko via Tarantool-patches
2021-07-26 23:45   ` Vladislav Shpilevoy via Tarantool-patches
2021-07-29 20:46     ` Sergey Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 11/16] box: make promote on the current leader a no-op Serge Petrenko via Tarantool-patches
2021-07-21 23:26   ` Vladislav Shpilevoy via Tarantool-patches
2021-07-23  7:45     ` Sergey Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 12/16] box: fix an assertion failure after a spurious wakeup in promote Serge Petrenko via Tarantool-patches
2021-07-21 23:29   ` Vladislav Shpilevoy via Tarantool-patches
2021-07-23  7:45     ` Sergey Petrenko via Tarantool-patches [this message]
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 13/16] box: allow calling promote on a candidate Serge Petrenko via Tarantool-patches
2021-07-15 14:06   ` Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 14/16] box: extract promote() settings to a separate method Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 15/16] replication: forbid implicit limbo owner transition Serge Petrenko via Tarantool-patches
2021-07-14 18:25 ` [Tarantool-patches] [PATCH v4 16/16] box: introduce `box.ctl.demote` Serge Petrenko via Tarantool-patches
2021-07-15 17:13   ` Serge Petrenko via Tarantool-patches
2021-07-15 20:11   ` [Tarantool-patches] [PATCH v4 17/16] replication: fix flaky election_qsync.test Serge Petrenko via Tarantool-patches
2021-07-26 23:43 ` [Tarantool-patches] [PATCH v4 00/16] forbid implicit limbo ownership transition Vladislav Shpilevoy via Tarantool-patches
2021-07-29 20:47   ` Sergey Petrenko via Tarantool-patches

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4f192cbf-5bad-4eb7-9ed0-2789d5d4c7d8@tarantool.org \
    --to=tarantool-patches@dev.tarantool.org \
    --cc=gorcunov@gmail.com \
    --cc=sergepetrenko@tarantool.org \
    --cc=v.shpilevoy@tarantool.org \
    --subject='Re: [Tarantool-patches] [PATCH v4 12/16] box: fix an assertion failure after a spurious wakeup in promote' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox