Re: invalid memory alloc request size

From: Gabriel Sánchez Martínez <gabrielesanchez(at)gmail(dot)com>
To: Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>, Postgres General <pgsql-general(at)postgresql(dot)org>
Subject: Re: invalid memory alloc request size
Date: 2014-12-10 19:50:32
Message-ID: 5488A408.7030306@gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general


On 12/10/2014 02:34 PM, Adrian Klaver wrote:
> On 12/10/2014 10:08 AM, Gabriel Sánchez Martínez wrote:
>>
>> On 12/10/2014 01:00 PM, Adrian Klaver wrote:
>>> On 12/10/2014 09:54 AM, Gabriel Sánchez Martínez wrote:
>>>>
>>>> On 12/10/2014 12:47 PM, Adrian Klaver wrote:
>>>>> On 12/10/2014 09:25 AM, Gabriel Sánchez Martínez wrote:
>>>>>>
>>>>>> On 12/10/2014 11:49 AM, Adrian Klaver wrote:
>>>>>>> On 12/10/2014 08:31 AM, Gabriel Sánchez Martínez wrote:
>>>>>>>>
>>>>>>>> On 12/10/2014 11:16 AM, Adrian Klaver wrote:
>>>>>>>>> On 12/10/2014 08:07 AM, Gabriel Sánchez Martínez wrote:
>>>>>>>>>> Hi all,
>>>>>>>>>>
>>>
>>>>>>>
>>>>>>> Also how did the Postgres server get installed?
>>>>>>
>>>>>> apt-get install
>>>>>
>>>>> Install from what repo?
>>>>
>>>> The default Ubuntu repository.
>>>>
>>>>>
>
>>
>>
>>
>>>
>>> The database was
>>>> upgraded to 9.3 when Ubuntu was upgraded to 14.04. But I ran it for a
>>>> while without issues. This issue is recent.
>>>
>>> So what was the version before?
>>
>> 9.1. It has always been whatever was on the Ubuntu repos.
>>
>>>
>>> Where I am going with this, is trying to determine whether you have a
>>> 'contaminated' data directory.
>>
>> I appreciate it! Perhaps it is worth mentioning that the server had
>> some issues with the linux md raid a while back. Two disks failed in
>> quick succession and some data was corrupted. Those disks have been
>> replaced and the RAID has been fine ever since. The database cluster
>> was recreated from scratch after that, from backups of a few months
>> before the disks started misbehaving. This is when that table was
>> created with pg_restore. There were no issues with the restore, and the
>> backup was from a few months before the disk issue, so I don't suspect
>> corruption in the backup.
>
> I would investigate the possibility that the raid is having problems
> again.

Everything looks good right now. I ran SMART checks on all drives and
there were no errors. I will restore the table from an earlier backup
and let you know if that takes care of the issue. Thanks for your help,
and please let me know if there is anything else I could do. I'll reply
to the list if there are any developments.

>
>>
>
>
>

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Jack Douglas 2014-12-10 20:23:31 Re: new index type with clustering in mind.
Previous Message Alvaro Herrera 2014-12-10 19:42:13 Re: new index type with clustering in mind.