[stgt] SEM Key Clash with Open-iSCSI initiator and LFS correction for ipc socket

FUJITA Tomonori fujita.tomonori at lab.ntt.co.jp
Sun Mar 6 06:45:01 CET 2011


On Thu, 3 Mar 2011 16:43:33 +0100
Roland Friedwagner <roland.friedwagner at wu.ac.at> wrote:

> Hello,
> 
> doing some regression test on a kvm cluster I had to run 
> the Open-iSCSI initiator (iscsid) and tgtd on the same machine.
> This brings up these to issues...
> 
> 
> 1. Logging of iscsd and tgtd broken
> 
> But this breaks logging of both iscsd and tgtd because they both
> run the logging semaphore with the same unix ipc key 0x000000a7
> (I think the reason is that the logging code from one project is
>  reused in the other project).
> The error messages found in syslog are:
>  tgtd: semop up failed
>  iscsid: semop down failed 22
> 
> 
> 2. IPC Namespace Socket in /tmp
> 
> I suggest strongly to move the ipc communication socket to /var/run
> to avoid the socket file is accidentally deleted by admins or 
> cron system maintenance scripts. 
> I think it is also more FHS conform this way.
> 
> 
> A test case for the breakage of logging facility can found in this
> redhat bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=676804
> 
> And this is my suggested patch to fix these issues:

Sounds reasonable.

But seems that your mailer corrupts the patch. I can't apply it. I
also need your signed-off-by. Please read 'Developer Notes' section in
the README file and resend the proper patch.


Thanks!
--
To unsubscribe from this list: send the line "unsubscribe stgt" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html



More information about the stgt mailing list