On Wed, Aug 3, 2016 at 8:46 AM Steve V <[email protected]> wrote:

>
> On Wednesday, August 3, 2016 at 12:38:47 AM UTC-5, Daniel Farina wrote:
>
>> On Tue, Aug 2, 2016 at 5:39 PM Steve V <[email protected]> wrote:
>>
>>> Does wal-e have any built in failure notification mechanisms, or is the
>>> best course of action to wrap the wal-e backup-push and wal-push commands
>>> in a shell script to catch and handle errors there?
>>>
>>
>> Sort of.
>>
>> One might use syslog.
>>
>> You might want to use your wrapper script, but be advised that for
>> wal-fetch in particular that it can be expected to fail, because Postgres
>> will probe for files, and uses the exit status for flow control.
>>
>
> This would only be used for wal-push, and backup-push. If I need to
> recover at any point I would be there managing the process, so I would be
> aware of any failures.
>

Alright, you could do that.  I think wal-push would be somewhat brutal as
it happens often and has retries.

For background information sake, when I have monitored WAL-E in the past, I
instead periodically count the number of `.ready` files in the
`archive_status` directory and the size of WAL on disk.

As for backups, one can check that a backup has been made within a certain
time period.

I think these are more robust measures than exit statuses, but your
situation may preclude such monitoring, for all I know.

-- 
You received this message because you are subscribed to the Google Groups 
"wal-e" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to