public inbox for pve-devel@lists.proxmox.com
 help / color / mirror / Atom feed
From: Fiona Ebner <f.ebner@proxmox.com>
To: "Proxmox VE development discussion" <pve-devel@lists.proxmox.com>,
	"Fabian Grünbichler" <f.gruenbichler@proxmox.com>
Subject: Re: [pve-devel] [PATCH v2 storage] lvm plugin: fix locking for rollback when using CLI
Date: Thu, 20 Nov 2025 14:35:17 +0100	[thread overview]
Message-ID: <b39fbaae-8e34-4cfe-85bf-fe529dd857d1@proxmox.com> (raw)
In-Reply-To: <1763644497.6m8gy24914.astroid@yuna.none>

Am 20.11.25 um 2:21 PM schrieb Fabian Grünbichler:
> On November 20, 2025 11:17 am, Fiona Ebner wrote:
>> @@ -1141,14 +1135,29 @@ my sub volume_snapshot_rollback_locked {
>>  sub volume_snapshot_rollback {
>>      my ($class, $scfg, $storeid, $volname, $snap) = @_;
>>  
>> -    return $class->cluster_lock_storage(
>> -        $storeid,
>> -        $scfg->{shared},
>> -        undef,
>> -        sub {
>> -            return volume_snapshot_rollback_locked($class, $scfg, $storeid, $volname, $snap);
>> -        },
>> -    );
>> +    my $cleanup_worker;
>> +
>> +    eval {
>> +        $class->cluster_lock_storage(
>> +            $storeid,
>> +            $scfg->{shared},
>> +            undef,
>> +            sub {
>> +                volume_snapshot_rollback_locked(
>> +                    $class, $scfg, $storeid, $volname, $snap, \$cleanup_worker,
>> +                );
>> +            },
>> +        );
>> +    };
>> +    my $err = $@;
>> +
>> +    # Spawn outside of the locked section, because with 'saferemove', the cleanup worker also needs
>> +    # to obtain the lock, and in CLI context, it will be awaited synchronously, see fork_worker().
>> +    fork_cleanup_worker($cleanup_worker);
> 
> I mean, this just worked because the forked cleanup sub waited long
> enough by chance for the lock/before acquiring the lock to allow the
> parent to release it in the non-CLI case, right? i.e., it was still
> wrong, because volume_snapshot_rollback_locked might have done more work
> after forking the cleanup worker while still holding the lock.. 

Yes, that is true.

> all the other code paths here have this pattern:
> 
> my $cleanup_worker = eval { lock and do something };
> fork_cleanup_worker($cleanup_worker)
> 
> which is exactly what this patch switches to, so consider this
> 
> Reviewed-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>

Thanks for the review!


_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

  reply	other threads:[~2025-11-20 13:35 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-11-20 10:17 Fiona Ebner
2025-11-20 13:21 ` Fabian Grünbichler
2025-11-20 13:35   ` Fiona Ebner [this message]
2025-11-20 14:06 ` [pve-devel] applied: " Thomas Lamprecht

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=b39fbaae-8e34-4cfe-85bf-fe529dd857d1@proxmox.com \
    --to=f.ebner@proxmox.com \
    --cc=f.gruenbichler@proxmox.com \
    --cc=pve-devel@lists.proxmox.com \
    /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
Service provided by Proxmox Server Solutions GmbH | Privacy | Legal