[163205] in North American Network Operators' Group

home help back first fref pref prev next nref lref last post

Re: Mailman reverting settings

daemon@ATHENA.MIT.EDU (Grant Ridder)
Fri May 24 21:04:30 2013

In-Reply-To: <CAPhg-wRGGzLh42uk1rwY7dHSP2wViKibBW8a4CtFZ0A3Oo9ZJw@mail.gmail.com>
Date: Fri, 24 May 2013 20:04:16 -0500
From: Grant Ridder <shortdudey123@gmail.com>
To: Phil Fagan <philfagan@gmail.com>
Cc: NANOG <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org

Hi,

I received a couple offlist replies.  To answer Phil's questions, iptables
appeared to have spiked the cpu to 100% and caused it to overload and
become unresponsive.  Var was lot filled up to my knowledge.

An offlist reply suggested that if the config.pck file gets corrupted, then
mailman will revert to using an old config.db file.  After doing a file
level restore of the appropriate config.pck file, the list returned to
normal.

Thanks,
Grant

On Fri, May 24, 2013 at 9:23 AM, Phil Fagan <philfagan@gmail.com> wrote:

> What hung the box? Core dump? Filled up var?
> On May 23, 2013 11:57 AM, "Grant Ridder" <shortdudey123@gmail.com> wrote:
>
>> Hi Everyone,
>>
>> Has anyone ever seen Mailman revert to an old user list?  This morning we
>> had out lists VM pounded on from India and hung the box.  After blocking
>> the ip on our firewall and rebooting the hung vm, everything came back up
>> except 1 list.  The list appears to have reverted to old settings.  I
>> found
>> the config.pck file and it does not have the current settings in it.  This
>> has happened once before.  The box is running Ubuntu 10.04 and Mailman
>> 2.1.13.  Anyone know what would cause this and how to fix it?  We are
>> working to try a file level restore from a backup.
>>
>> Thanks,
>> Grant
>>
>> ~~~~~~~~~~
>>
>> Grant Ridder
>>
>> Assistant Systems Administrator
>>
>> Milwaukee School of Engineering <http://www.msoe.edu/>
>>
>

home help back first fref pref prev next nref lref last post