[PATCH] CTDB fixups

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

[PATCH] CTDB fixups

Samba - samba-technical mailing list
* ctdb-common: Ensure unused bytes in union are initialised

  Initialised data can make it to ctdbd.  We should probably fix this
  in smbd too.

* ctdb-tests: Have fake daemon log when it parses public IPs

  Other initialisation is logged, so the absence here was confusing.

* ctdb-scripts: Don't bother checking PID file when starting ctdbd

  From commit message:

    This is an optimisation that can cause incorrect results.  If ctdbd
    was killed and there is a stale PID file then this will often cause
    "CTDB exited during initialisation".  The wrapper reads the old PID
    from the PID file, finds the PID gone, complains and exits.

    It is better to drop this code and finally get this right.  If
    ctdbd does exit early then it will take CTDB_STARTUP_TIMEOUT
    (default 10) seconds before the wrapper fails.  That's not too
    bad...

* ctdb-common: Fix stale socket removal

  Fix a regression in master.  Patch by Amitay, already has my review.

* ctdb-tests: Add sock_daemon test for stale socket handling

  Add a new test to avoid repeat of regression.

Please review and maybe push...

peace & happiness,
martin

ctdb.patch (16K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: [PATCH] CTDB fixups

Samba - samba-technical mailing list
On Mon, Nov 6, 2017 at 12:38 PM, Martin Schwenke via samba-technical <
[hidden email]> wrote:

> * ctdb-common: Ensure unused bytes in union are initialised
>
>   Initialised data can make it to ctdbd.  We should probably fix this
>   in smbd too.
>
> * ctdb-tests: Have fake daemon log when it parses public IPs
>
>   Other initialisation is logged, so the absence here was confusing.
>
> * ctdb-scripts: Don't bother checking PID file when starting ctdbd
>
>   From commit message:
>
>     This is an optimisation that can cause incorrect results.  If ctdbd
>     was killed and there is a stale PID file then this will often cause
>     "CTDB exited during initialisation".  The wrapper reads the old PID
>     from the PID file, finds the PID gone, complains and exits.
>
>     It is better to drop this code and finally get this right.  If
>     ctdbd does exit early then it will take CTDB_STARTUP_TIMEOUT
>     (default 10) seconds before the wrapper fails.  That's not too
>     bad...
>
> * ctdb-common: Fix stale socket removal
>
>   Fix a regression in master.  Patch by Amitay, already has my review.
>
> * ctdb-tests: Add sock_daemon test for stale socket handling
>
>   Add a new test to avoid repeat of regression.
>
> Please review and maybe push...
>
> peace & happiness,
> martin
>

Pushed to autobuild.

Amitay.