Mailing List Archive

Replag
Hi,

it looks like there is a very big replication lag in dewiki.

Is a debugging already in progress?

Kind regards,
Martin ...
Re: Replag [ In reply to ]
Yup, see https://phabricator.wikimedia.org/T337446.

Zabe

> Am 28.05.2023 um 22:46 schrieb Martin Domdey <dr.taxon@gmail.com>:
>
> ?
> Hi,
>
> it looks like there is a very big replication lag in dewiki.
>
> Is a debugging already in progress?
>
> Kind regards,
> Martin ...
> _______________________________________________
> Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org
> To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org
> https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
Re: Replag [ In reply to ]
Hello,

I am trying to get this fixed as fast as I can.

There will be brief periods of total unavailability for a section at the
time. Right now s7 is totally down (the data was almost a week out of date
anyways).
Next will be s5.

Sorry for the inconveniences.
Manuel.

On Sun, May 28, 2023 at 11:04?PM Alec Vorwerk <alec@vc-celle.de> wrote:

> Yup, see https://phabricator.wikimedia.org/T337446.
>
> Zabe
>
> Am 28.05.2023 um 22:46 schrieb Martin Domdey <dr.taxon@gmail.com>:
>
> ?
> Hi,
>
> it looks like there is a very big replication lag in dewiki.
>
> Is a debugging already in progress?
>
> Kind regards,
> Martin ...
> _______________________________________________
> Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org
> To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org
> https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
>
> _______________________________________________
> Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org
> To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org
> https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
Re: Replag [ In reply to ]
Hello,

Cross posting my latest update on the task:
https://phabricator.wikimedia.org/T337446#8894523



*I am reducing the priority of this as all the hosts have been recloned now
and data should be up to date.We shouldn't be surprised if s6 and s8 (the
sections that never break) end up breaking on the sanitarium hosts, as if
the problem was 10.4.29, data might have been corrupted there and simply
didn't show up yet.I am going to do some data checking now on the recloned
versions before closing this task, hopefully by Monday if everything goes
fine in the next few days.*

*Things might still be slow on some of the tools as we are adding the
special indexes used in wikireplicas, that can be tracked at T337734
<https://phabricator.wikimedia.org/T337734>*

Sorry for the inconvenience this has caused (and the slowness we are still
experiencing)
Manuel.

On Mon, May 29, 2023 at 3:59?PM Manuel Arostegui <marostegui@wikimedia.org>
wrote:

> Hello,
>
> I am trying to get this fixed as fast as I can.
>
> There will be brief periods of total unavailability for a section at the
> time. Right now s7 is totally down (the data was almost a week out of date
> anyways).
> Next will be s5.
>
> Sorry for the inconveniences.
> Manuel.
>
> On Sun, May 28, 2023 at 11:04?PM Alec Vorwerk <alec@vc-celle.de> wrote:
>
>> Yup, see https://phabricator.wikimedia.org/T337446.
>>
>> Zabe
>>
>> Am 28.05.2023 um 22:46 schrieb Martin Domdey <dr.taxon@gmail.com>:
>>
>> ?
>> Hi,
>>
>> it looks like there is a very big replication lag in dewiki.
>>
>> Is a debugging already in progress?
>>
>> Kind regards,
>> Martin ...
>> _______________________________________________
>> Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org
>> To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org
>>
>> https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
>>
>> _______________________________________________
>> Wikitech-l mailing list -- wikitech-l@lists.wikimedia.org
>> To unsubscribe send an email to wikitech-l-leave@lists.wikimedia.org
>>
>> https://lists.wikimedia.org/postorius/lists/wikitech-l.lists.wikimedia.org/
>
>