Tarantool development patches archive
 help / color / mirror / Atom feed
From: Konstantin Osipov <kostja@tarantool.org>
To: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: tarantool-patches@freelists.org
Subject: Re: [PATCH v2 1/6] xlog: store prev vclock in xlog header
Date: Thu, 5 Jul 2018 14:22:28 +0300	[thread overview]
Message-ID: <20180705112228.GA23856@chai> (raw)
In-Reply-To: <20180705082351.qgqohxywjnc2jttp@esperanza>

* Vladimir Davydov <vdavydov.dev@gmail.com> [18/07/05 14:01]:
> On Thu, Jul 05, 2018 at 09:52:41AM +0300, Konstantin Osipov wrote:
> > * Vladimir Davydov <vdavydov.dev@gmail.com> [18/06/29 19:49]:
> > > This patch adds a new key to xlog header, PrevVclock, which contains the
> > > vclock of the previous xlog file in the xlog directory. It is set by
> > > xdir_create_xlog() to the last vclock in xdir::index. The new key is
> > > only present in XLOG files (it doesn't make sense for SNAP or VYLOG
> > > anyway). It will be used to make the check for xlog gaps more thorough.
> > 
> > What about the following diff:
> > 
> > ---
> > diff --git a/src/box/xlog.c b/src/box/xlog.c
> > index 59459b25d..a60f2ffa8 100644
> > --- a/src/box/xlog.c
> > +++ b/src/box/xlog.c
> > @@ -126,12 +126,10 @@ xlog_meta_format(const struct xlog_meta *meta, char *buf, int size)
> >  		VCLOCK_KEY ": %s\n",
> >  		meta->filetype, v13, PACKAGE_VERSION, instance_uuid, vstr);
> >  	free(vstr);
> > -	if (meta->has_prev_vclock) {
> > -		vstr = vclock_to_string(&meta->prev_vclock);
> > -		SNPRINT(total, snprintf, buf, size,
> > -			PREV_VCLOCK_KEY ": %s\n", vstr);
> > -		free(vstr);
> > -	}
> > +	vstr = vclock_to_string(&meta->prev_vclock);
> > +	SNPRINT(total, snprintf, buf, size,
> > +		PREV_VCLOCK_KEY ": %s\n", vstr);
> > +	free(vstr);
> >  	SNPRINT(total, snprintf, buf, size, "\n");
> >  	assert(total > 0);
> >  	return total;
> > @@ -263,7 +261,6 @@ xlog_meta_parse(struct xlog_meta *meta, const char **data,
> >  			 */
> >  			if (parse_vclock(val, val_end, &meta->prev_vclock) != 0)
> >  				return -1;
> > -			meta->has_prev_vclock = true;
> 
> We need to know if PrevVclock is available so that we can skip the check
> when recovering from an xlog generated by a previous version.

In theory we could use Version key for this.

> 
> >  		} else if (memcmp(key, VERSION_KEY, key_end - key) == 0) {
> >  			/* Ignore Version: for now */
> >  		} else {
> > @@ -926,12 +923,10 @@ xdir_create_xlog(struct xdir *dir, struct xlog *xlog,
> >  	 * For WAL dir: store vclock of the previous xlog file
> >  	 * to check for gaps on recovery.
> >  	 */
> > -	if (dir->type == XLOG && !vclockset_empty(&dir->index)) {
> > +	if (!vclockset_empty(&dir->index)) {
> >  		vclock_copy(&meta.prev_vclock, vclockset_last(&dir->index));
> > -		meta.has_prev_vclock = true;
> >  	} else {
> >  		vclock_create(&meta.prev_vclock);
> > -		meta.has_prev_vclock = false;
> >  	}
> 
> Storing vclock in snap and vylog files doesn't make any sense IMHO.
> I'd prefer to check dir->type explicitly.
> 
> If you dislike the extra variable, we can probably introduce a helper to
> check if vclock is set, something like
> 
> 	static inline bool
> 	vclock_is_set(const struct vclock *vclock)
> 	{
> 		return vclock->signature >= 0;
> 	}

This is better IMHO.


-- 
Konstantin Osipov, Moscow, Russia, +7 903 626 22 32
http://tarantool.io - www.twitter.com/kostja_osipov

  reply	other threads:[~2018-07-05 11:22 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-29 16:48 [PATCH v2 0/6] Create empty xlog on shutdown Vladimir Davydov
2018-06-29 16:48 ` [PATCH v2 1/6] xlog: store prev vclock in xlog header Vladimir Davydov
2018-07-05  6:49   ` Konstantin Osipov
2018-07-05  6:52   ` Konstantin Osipov
2018-07-05  8:23     ` Vladimir Davydov
2018-07-05 11:22       ` Konstantin Osipov [this message]
2018-07-10 16:28         ` [PATCH] xlog: get rid of xlog_meta::has_prev_vclock Vladimir Davydov
2018-06-29 16:48 ` [PATCH v2 2/6] xlog: differentiate between closed and never opened cursor Vladimir Davydov
2018-06-29 16:48 ` [PATCH v2 3/6] recovery: make LSN gap check more thorough Vladimir Davydov
2018-06-29 16:48 ` [PATCH v2 4/6] recovery: promote recovery clock even if the WAL is empty Vladimir Davydov
2018-06-29 16:48 ` [PATCH v2 5/6] wal: create empty xlog on shutdown Vladimir Davydov
2018-06-29 16:48 ` [PATCH v2 6/6] error: move XlogGapError to box/error.h Vladimir Davydov

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=20180705112228.GA23856@chai \
    --to=kostja@tarantool.org \
    --cc=tarantool-patches@freelists.org \
    --cc=vdavydov.dev@gmail.com \
    --subject='Re: [PATCH v2 1/6] xlog: store prev vclock in xlog header' \
    /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