Subject: Re: Health checks after machine crash?
To: "pgsql-admin@postgresql.org" <pgsql-admin@postgresql.org>
References: <b1aff4d1-8db0-77bf-3eb9-3f181630cf61@gmail.com>
 <87y3hd4xfc.fsf@jsievers.enova.com>
From: Ron <ronljohnsonjr@gmail.com>
Message-ID: <14e7c714-f747-05be-9a80-4066b607a6a2@gmail.com>
Date: Mon, 23 Apr 2018 14:09:33 -0500
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.7.0
MIME-Version: 1.0
In-Reply-To: <87y3hd4xfc.fsf@jsievers.enova.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Precedence: bulk


On 04/23/2018 12:58 PM, Jerry Sievers wrote:
> Ron <ronljohnsonjr@gmail.com> writes:
>
>> Hi,
>>
>> v8.4 (Yes, it's archaic.  Upgrade plan is in progress.)
> Please do!  :-)
>
>> What queries should I run against the catalog tables to see whether or
>> not the cluster is healthy after reboot?
> Is the cluster confi'd to be at risk of corruption after crash.. such as
> full_page_write=off, fsync=off?

#full_page_writes = on
fsync = on

> Do your logs show any obvious sign of corruption, bad blocks etc?

Nothing I've notived.

> You may have little to worry about since Pg is very crash tolerant even
> old versions 8.4 and prior.
>
> I'd try vacuum and analyze whole cluster plus dumpall >/dev/null and
> check for errors there too.  Any sort of corruption likely to cause
> application problems likely to show up in so doing.

Will do.


-- 
Angular momentum makes the world go 'round.