Re: Unable to recovery due missing wal_file

From: Martín Marqués <martin(at)2ndquadrant(dot)com>
To: Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>, Patrick B <patrickbakerbr(at)gmail(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Unable to recovery due missing wal_file
Date: 2016-07-04 23:37:34
Message-ID: f72572aa-fd8c-87e8-69d8-bfc7e25f05d1@2ndquadrant.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

El 04/07/16 a las 01:06, Adrian Klaver escribió:
> On 07/03/2016 06:21 PM, Patrick B wrote:
>>
>>
>> Not sure that would have mattered for the reasons below.
>>
>> You might want to take a look at the below:
>>
>> https://www.postgresql.org/docs/9.5/static/wal.html
>>
>> In particular:
>>
>> https://www.postgresql.org/docs/9.5/static/wal-intro.html
>>
>> Short version WAL files are essential to restoring and on the
>> originating server are recycled, subject to the configuration
>> parameters explained here:
>>
>> https://www.postgresql.org/docs/9.5/static/wal-configuration.html
>>
>> So the wal archiving you have set up is not storing everything, it
>> removes older files over time?
>>
>>
>> Yes... it removes... I've changed to store them for up to 72h.
>
> You say it took 10 days to run the pg_basebackup, so I am not keeping
> the last 72 hrs is going to help.

Not only that, if you add up another 500GB to transfer over the WALs,
that might mean another 2 to 3 days to finish the transfer (and that if
it's only 500GB of WALs)

BTW, 2TB in 10 days means an avg speed of 2.3Mb/s. I guess this must be
some standby in a DR site without a dedicated network bandwidth.

Wouldn't it be faster to clone the disk locally, detach it and send it
over with a Courier? ;)

Regards,

--
Martín Marqués http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Haribabu Kommi 2016-07-05 06:39:02 Re: Question about "grant create on database" and pg_dump/pg_dumpall
Previous Message Nicholson, Brad (Toronto, ON, CA) 2016-07-04 20:03:40 9.6 Beta 2 Performance Regression on Recursive CTE