Tarantool development patches archive
 help / color / mirror / Atom feed
From: Cyrill Gorcunov <gorcunov@gmail.com>
To: Konstantin Osipov <kostja.osipov@gmail.com>
Cc: tml <tarantool-patches@dev.tarantool.org>
Subject: Re: [Tarantool-patches] [PATCH v10 4/4] test: add replication/applier-rollback
Date: Sat, 15 Feb 2020 20:59:53 +0300	[thread overview]
Message-ID: <20200215175953.GC2527@uranus> (raw)
In-Reply-To: <20200215173851.GD19902@atlas>

On Sat, Feb 15, 2020 at 08:38:51PM +0300, Konstantin Osipov wrote:
> * Cyrill Gorcunov <gorcunov@gmail.com> [20/02/14 17:06]:
> >  STRS(applier_state, applier_STATE);
> >  
> > @@ -832,6 +833,17 @@ applier_apply_tx(struct stailq *rows)
> >  	trigger_create(on_commit, applier_txn_commit_cb, NULL, NULL);
> >  	txn_on_commit(txn, on_commit);
> >  
> > +	/*
> > +	 * FIXME: Move this injection somewhere inside
> > +	 * txn_write, but since it has own bug (gh 4776)
> > +	 * we will cure it a bit later.
> > +	 */
> > +	ERROR_INJECT(ERRINJ_REPLICA_TXN_WRITE, {
> > +		diag_set(ClientError, ER_INJECTION,
> > +			 "replica txn write injection");
> > +		goto rollback;
> > +	});
> 
> I thought yo wrote in the changes in v10 that you pushed the error
> down into txn layer. But here you don't do it. What am I missing?

Not yet. Pushing it sown cause https://github.com/tarantool/tarantool/issues/4776
I plan to fix it and then get back to this issue. Look this
all code is still a bit vague for me, so I've to understand every
byte before doing some more deep changes. For exactly this reason
the patch in the series does NOT change the current logic of
code flow much but tries to leave everything as it were except
a hot fix for nil dereference.

That said, Kostya, I remember about your proposal and FIXME
left here for exactly that.

  reply	other threads:[~2020-02-15 17:59 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-14 14:03 [Tarantool-patches] [PATCH v10 0/4] box/replication: add missing diag set and fix sigsegv Cyrill Gorcunov
2020-02-14 14:03 ` [Tarantool-patches] [PATCH v10 1/4] box/request: add missing OutOfMemory diag_set Cyrill Gorcunov
2020-02-15 17:33   ` Konstantin Osipov
2020-02-14 14:03 ` [Tarantool-patches] [PATCH v10 2/4] box/applier: add missing diag_set on region_alloc failure Cyrill Gorcunov
2020-02-15 17:34   ` Konstantin Osipov
2020-02-14 14:03 ` [Tarantool-patches] [PATCH v10 3/4] box/applier: prevent nil dereference on applier rollback Cyrill Gorcunov
2020-02-15 17:37   ` Konstantin Osipov
2020-02-15 18:24     ` Cyrill Gorcunov
2020-02-15 18:51       ` Konstantin Osipov
2020-02-14 14:03 ` [Tarantool-patches] [PATCH v10 4/4] test: add replication/applier-rollback Cyrill Gorcunov
2020-02-15 17:38   ` Konstantin Osipov
2020-02-15 17:59     ` Cyrill Gorcunov [this message]
2020-02-14 14:12 ` [Tarantool-patches] [PATCH v10 0/4] box/replication: add missing diag set and fix sigsegv Cyrill Gorcunov

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=20200215175953.GC2527@uranus \
    --to=gorcunov@gmail.com \
    --cc=kostja.osipov@gmail.com \
    --cc=tarantool-patches@dev.tarantool.org \
    --subject='Re: [Tarantool-patches] [PATCH v10 4/4] test: add replication/applier-rollback' \
    /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