[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: cfservd dying
From: |
Mark . Burgess |
Subject: |
Re: cfservd dying |
Date: |
Thu, 13 Mar 2003 07:41:28 +0100 (MET) |
Just get cfengine to restart it .. but, if this is happening a lot, a
gdb trace would be useful so that the bug can be fixed. I have not
seen cfservd crashing for almost a year.
M
On 12 Mar, Ryan Anderson wrote:
> While I haven't done this with cfservd yet, you could wrap cfservd in
> /etc/inittab with the 'respawn' option so that it will respawn the
> process if it ever dies. Anyone try this with cfservd? Any problems with
> it?
>
> RCA
> --
> Ryan C. Anderson
> Unix Administrator
> United Defense L.P.
> desk 763.572.6684
> pager 952.235.9936
> mobile 612.419.9362
>
>>>> Yves <Y99@gmx.de> 03/12/03 04:15PM >>>
>>Subject: cfservd dying
>>From: paguerlais airfrance.fr
>>Date: Wed, 11 Sep 2002 17:30:57 +0200
>>Newsgroups: gmane.comp.sysutils.cfengine.general
>>Sensitivity:
>>
>>
>>Hi all,
>>
>>it seems to me that somebody already mailed about that, but I can't
> find
>>out again that damned thread ;-)
>>
>>I'm running cfengine 2.0.4 on a RedHat 7.1 box, and trying to use
> cfrun
>>from the same box to execute its cfagent. cfservd starts the
> connection
>>with cfrun, but dies somewhat later. This happens only when cfrun is
>>executed on the same box than the cfservd it must contact.
>>If I run cfservd in debug mode, I can see that the authentication
> process
>>seems ok, but thereafter cfservd dies with the following messages :
>>
>>IPV4 address
>>sockaddr_ntop()
>>Address family was 1
>>my_host.my_domain::Software failure in sockaddr_ntop
>>
>>ReleaseCurrentLock(cfservd)
>>failed
>>
>>
>>Does somebody has already seen this ?
>>
>>Attn Mark : you used the same message (software failure in
> sockaddr_ntop)
>>in two functions, sockaddr_ntop and sockaddr_pton. Perhaps you've
> forgotten
>>to change the function name for the second one ?
>>
>>Thanks
>>Patrice
>
> Hi Patrice,
>
> I have the same problem. My cfservd dies very often.
> I tried (for the first time) "cfservd -d 2" on cfengine master:
>
>
> cfservd Debug mode: running in foreground
> GetNameInfo()
> AddClassToHeap(aix)
> Appending [aix]
> ....
> Host IPs from NAT which we don't verify :
>
> IP: xx.xx.82.215
> Dynamical Host IPs (e.g. DHCP) whose bindings could vary over time :
>
> Address family was 0
> cfmaster::Software failure in sockaddr_ntop
>
> ReleaseCurrentLock(cfservd)
>
>
> and cfservd doesn't start.
>
> If I start cfservd without options it works and ca. 800 clients contact
> the cfengine master every half an hour. But as mentioned above the
> cfservd dies very often (ca. every other day) without any hints in
> SYSLOG.
>
> How can I analyze it ?
> Does anybody have an idea ? Is it an compiling problem ?
>
> basic conditions:
> AIX 4.3.3
> cfengine v2.0.4
>
> Gruß Yves
>
> Y99@gmx.de am 12.03.2003
>
>
> _______________________________________________
> Help-cfengine mailing list
> Help-cfengine@gnu.org
> http://mail.gnu.org/mailman/listinfo/help-cfengine
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Work: +47 22453272 Email: Mark.Burgess@iu.hio.no
Fax : +47 22453205 WWW : http://www.iu.hio.no/~mark
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
- Re: cfservd dying, Yves, 2003/03/12
- Re: cfservd dying, Ryan Anderson, 2003/03/12
- Re: cfservd dying,
Mark . Burgess <=
- RE: cfservd dying, Wheeler, John, 2003/03/13
- Re: cfservd dying, Yves, 2003/03/13
- Re: cfservd dying, Yves, 2003/03/14