couchdb-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Robert Newson <rnew...@apache.org>
Subject Re: Problem with restart of continuous replication
Date Tue, 23 Apr 2013 09:30:19 GMT
Calle,

I would set the retry limit to infinity, I think you'll agree with me
that a persistent replication that gives up is hardly persistent at
all;

curl -X PUT host:port/_config/replicator/max_replication_retry_count
-d '"infinity"'

Note that that's a double-quoted string inside single quotes to cope
with shell escaping.

B.

On 23 April 2013 10:25, Dave Cottlehuber <dch@jsonified.com> wrote:
> On 23 April 2013 11:05, Calle Arnesten <calle.arnesten@codekick.com> wrote:
>> Hi Dave,
>>
>>> replication will retry a certain number of times and then give up if
>>> the connection does not work, 10 by default.
>>>
>>> See http://wiki.apache.org/couchdb/Replication#New_features_introduced_in_CouchDB_1.2.0
>>> for the details.
>>
>> Nice. So if I bumped the retries_per_request parameter to say 100, it would have
managed the 2 hour disconnect and then it would have restarted?
>
> Maybe, depends how quickly the failures add up.
>
>>> use the replicator DB mentioned in the above link, and monitor the
>>> state of that named document:
>>>
>>> GET /_replicator/my_replication_task
>
> Bob & I think _active_tasks is your best bet.
>
> A+
> Dave

Mime
View raw message