From: Christophe Varoqui <christophe.varoqui@gmail.com>
To: Hannes Reinecke <hare@suse.de>
Cc: dm-devel@redhat.com
Subject: Re: [PATCH 42/42] multipathd: lock vectors during initial configuration
Date: Wed, 09 Jan 2013 00:37:22 +0100 [thread overview]
Message-ID: <1357688242.12137.3.camel@lapoo.opensvc.com> (raw)
In-Reply-To: <1357653259-62650-42-git-send-email-hare@suse.de>
On mar., 2013-01-08 at 14:54 +0100, Hannes Reinecke wrote:
> During initial configuration the CLI thread is already running,
> so we need to lock the vectors here to not race with the
> 'reconfigure' CLI command.
>
Are you sure of the 41 - 42 patches ?
- 41 moves the vecs locking from reconfigure caller to reconfigure
- 42 removes the locking from reconfigure
Regards,
Christophe Varoqui
www.opensvc.com
> Signed-off-by: Hannes Reinecke <hare@suse.de>
> ---
> multipathd/main.c | 4 +---
> 1 files changed, 1 insertions(+), 3 deletions(-)
>
> diff --git a/multipathd/main.c b/multipathd/main.c
> index 395307e..8917499 100644
> --- a/multipathd/main.c
> +++ b/multipathd/main.c
> @@ -1389,7 +1389,6 @@ reconfigure (struct vectors * vecs)
> struct config * old = conf;
> int retval = 1;
>
> - lock(vecs->lock);
> /*
> * free old map and path vectors ... they use old conf state
> */
> @@ -1410,7 +1409,6 @@ reconfigure (struct vectors * vecs)
> retval = 0;
> }
>
> - unlock(vecs->lock);
> return retval;
> }
>
> @@ -1641,9 +1639,9 @@ child (void * param)
> /*
> * fetch and configure both paths and multipaths
> */
> - lock(vecs->lock);
> running_state = DAEMON_CONFIGURE;
>
> + lock(vecs->lock);
> if (configure(vecs, 1)) {
> unlock(vecs->lock);
> condlog(0, "failure during configuration");
next prev parent reply other threads:[~2013-01-08 23:37 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-08 13:53 [PATCH 00/42] SLES resync Hannes Reinecke
2013-01-08 13:53 ` [PATCH 01/42] libmultipath: Invalid check for mpp->wwid in dm_addmap() Hannes Reinecke
2013-01-08 13:53 ` [PATCH 02/42] Remove newline from condlog() Hannes Reinecke
2013-01-08 13:53 ` [PATCH 03/42] Fixup pathgroup allocation in disassemble_map() Hannes Reinecke
2013-01-08 13:53 ` [PATCH 04/42] libmultipath: resource leak in read_value_block() Hannes Reinecke
2013-01-08 13:53 ` [PATCH 05/42] prio: fix merging of prioritizers with different args Hannes Reinecke
2013-01-08 13:53 ` [PATCH 06/42] Accept several whitespaces in bindings file Hannes Reinecke
2013-01-08 13:53 ` [PATCH 07/42] Add TAGS makefile target Hannes Reinecke
2013-01-08 13:53 ` [PATCH 08/42] libmultipath: Fix typo in mp_prio_handler() Hannes Reinecke
2013-01-08 13:53 ` [PATCH 09/42] Do not trigger a map reload on priority updates Hannes Reinecke
2013-01-08 13:53 ` [PATCH 10/42] Introduce MP_FAST_IO_FAIL_UNSET Hannes Reinecke
2013-01-08 13:53 ` [PATCH 11/42] Checker name is not displayed on failure Hannes Reinecke
2013-01-08 13:53 ` [PATCH 12/42] Valgrind fixes for prioritizer Hannes Reinecke
2013-01-08 13:53 ` [PATCH 13/42] Incorrect inquiry vendor length in hds prioritizer Hannes Reinecke
2013-01-08 13:53 ` [PATCH 14/42] Print out multipath alias for flush_on_last_del messages Hannes Reinecke
2013-01-08 13:53 ` [PATCH 15/42] Clarify setting origin in propsel.c Hannes Reinecke
2013-01-08 13:53 ` [PATCH 16/42] libmultipath: error checking in remove_features() Hannes Reinecke
2013-01-08 13:53 ` [PATCH 17/42] Increase parameter buffer Hannes Reinecke
2013-01-08 13:53 ` [PATCH 18/42] Check return code from pathinfo() Hannes Reinecke
2013-01-08 13:53 ` [PATCH 19/42] Inconsistent string quoting Hannes Reinecke
2013-01-08 13:53 ` [PATCH 20/42] Switch off 'queue_if_no_path' before removing maps Hannes Reinecke
2013-01-08 13:53 ` [PATCH 21/42] Double free in disassemble_map() Hannes Reinecke
2013-01-08 13:54 ` [PATCH 22/42] libmultipath: prio keyword ignored for multipath config Hannes Reinecke
2013-01-08 13:54 ` [PATCH 23/42] Path checker should return PATH_DOWN when no path is found Hannes Reinecke
2013-01-08 13:54 ` [PATCH 24/42] Do not call sysfs_get_timeout for non-SCSI devices Hannes Reinecke
2013-01-08 13:54 ` [PATCH 25/42] Make log_pthread more robust Hannes Reinecke
2013-01-09 0:16 ` Christophe Varoqui
2013-01-09 19:15 ` Xose Vazquez Perez
2013-01-08 13:54 ` [PATCH 26/42] Print log messages when updating tables failed Hannes Reinecke
2013-01-08 13:54 ` [PATCH 27/42] Update 'no_path_retry' correctly for failed paths Hannes Reinecke
2013-01-08 13:54 ` [PATCH 28/42] Clean up uevent queue on shutdown Hannes Reinecke
2013-01-08 13:54 ` [PATCH 29/42] libmultipath: Print out uevent sequence number Hannes Reinecke
2013-01-08 13:54 ` [PATCH 30/42] Fix race condition in stop_waiter_thread() Hannes Reinecke
2013-01-08 13:54 ` [PATCH 31/42] Use VECTOR_SIZE() defines Hannes Reinecke
2013-01-08 13:54 ` [PATCH 32/42] Make 'allocated' an integer in vector.h Hannes Reinecke
2013-01-08 13:54 ` [PATCH 33/42] Syntax error in /etc/init.d/boot.multipath Hannes Reinecke
2013-01-08 13:54 ` [PATCH 34/42] multipath.init.suse: Update usage message Hannes Reinecke
2013-01-08 13:54 ` [PATCH 35/42] multipath.conf.5: Clarify dev_loss_tmo settings Hannes Reinecke
2013-01-08 13:54 ` [PATCH 36/42] Clarify dev_loss_tmo capping in multipath.conf.5 Hannes Reinecke
2013-01-08 13:54 ` [PATCH 38/42] multipathd: Ignore errors when creating pidfile Hannes Reinecke
2013-01-08 13:54 ` [PATCH 39/42] multipathd deadlocks during restart Hannes Reinecke
2013-01-08 13:54 ` [PATCH 40/42] multipathd: sighandlers might use uninitialized gvecs Hannes Reinecke
2013-01-08 13:54 ` [PATCH 41/42] multipathd: crash in reconfigure CLI command Hannes Reinecke
2013-01-08 13:54 ` [PATCH 42/42] multipathd: lock vectors during initial configuration Hannes Reinecke
2013-01-08 23:37 ` Christophe Varoqui [this message]
2013-01-09 7:01 ` Hannes Reinecke
2013-01-08 23:53 ` [PATCH 00/42] SLES resync Christophe Varoqui
2013-01-10 5:36 ` Benjamin Marzinski
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=1357688242.12137.3.camel@lapoo.opensvc.com \
--to=christophe.varoqui@gmail.com \
--cc=christophe.varoqui@opensvc.com \
--cc=dm-devel@redhat.com \
--cc=hare@suse.de \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox