Switch Datacenter

See also the blog posts from the "2016 Failover" and "2017 Failover" on the Wikimedia Blog.

Introduction

A datacenter switchover (from eqiad to codfw, or vice-versa) comprises switching over multiple different components, some of which can happen independently and many of which need to happen in lockstep. This page documents all the steps needed to switch over from a master datacenter to another one, broken up by component.

Schedule for 2020 switch

• Services: Monday, August 31st, 2020 14:00 UTC
• Traffic: Monday, August 31st, 2020 15:00 UTC
• MediaWiki: Tuesday, September 1st, 2020 14:00 UTC

Switching back:

• Traffic: Thursday, September 17th, 2020 17:00 UTC
• MediaWiki: Tuesday, October 27th, 2020 14:00 UTC
• Services: Wednesday, October 28th, 2020 14:00 UTC

Per-service switchover instructions

MediaWiki

We divide the process in logical phases that should be executed sequentially. Within any phase, top-level tasks can be executed in parallel to each other, while subtasks are to be executed sequentially to each other. The phase number is referred to in the names of the tasks in the operations/cookbooks repository, in the cookbooks/sre/switchdc/mediawiki/ path.

Days in advance preparation

1. OPTIONAL: SKIP IN AN EMERGENCY: Make sure databases are in a good state. Normally this requires no operation, as the passive datacenter databases are always prepared to receive traffic, so there are no actionables. Some things that #DBAs normally should make sure for the most optimal state possible (sanity checks):
   • There is no ongoing long-running maintenance that affects database availability or lag (schema changes, upgrades, hardware issues, etc.). Depool those servers that are not ready.
   • Replication is flowing from eqiad -> codfw and from codfw -> eqiad (sometimes replication gets stopped in the passive -> active direction to facilitate maintenance)
   • All database servers have its buffer pool filled up. This is taken care automatically with the automatic buffer pool warmup functionality. For sanity checks, some sample load could be sent to the mediawiki application server to check requests happen as quickly as in the active datacenter.
   • These where the things we prepared/checked for the 2018 switch
2. Make absolutely sure that parsercache replication is working from the active to the passive DC. This is important

Phase 0 - preparation

1. Disable puppet on maintenance hosts in both eqiad and codfw: 00-disable-puppet.py
2. Reduce the TTL on appservers-ro, appservers-rw, api-ro, api-rw, jobrunner, videoscaler, parsoid-php to 10 seconds: 00-reduce-ttl.py The warmup step (below) typically takes longer than the old TTL (5 minutes) so this step doesn't wait it to expire. Make sure that at least 5 minutes have passed before moving to Phase 1.
3. Warm up APC running the mediawiki-cache-warmup on the new site clusters. The warmup queries are automatically sent three times, because empircally that's when the response times stabilize: 00-warmup-caches.py
   • The global warmup against the appservers cluster
   • The apc-warmup against all hosts in the appservers cluster.
4. Set downtime for Read only checks on mariadb masters changed on Phase 3 (they will complain after the etcd primary master changes and until puppet runs there)

Phase 1 - stop maintenance

1. Stop maintenance jobs in the active site and kill all the cronjobs on the maintenance host in the active site: 01-stop-maintenance.py

Phase 2 - read-only mode

1. Go to read-only mode by changing the ReadOnly conftool value: 02-set-readonly.py

Phase 3 - lock down database masters

1. Put old-site core DB masters (shards: s1-s8, x1, es4-es5) in read-only mode and wait for the new site's databases to catch up replication: 03-set-db-readonly.py

Phase 4 - switch active datacenter configuration

1. Switch the discovery records and MediaWiki active datacenter: 04-switch-mediawiki.py
   • Flip appservers-ro, appservers-rw, api-ro, api-rw, jobrunner, videoscaler, parsoid-php to pooled=true in the new site. Since both sites are now pooled in etcd, this will not actually change the DNS records for the active datacenter.
   • Flip WMFMasterDatacenter from the old site to the new.
   • Flip appservers-ro, appservers-rw, api-ro, api-rw, jobrunner, videoscaler, parsoid-php to pooled=false in the old site. After this, DNS will be changed for the old DC and internal applications (except mediawiki) will start hitting the new DC

Phase 5 - Invert Redis replication for MediaWiki sessions

1. Invert the Redis replication for the sessions cluster: 05-invert-redis-sessions.py

Phase 6 - Set new site's databases to read-write

1. Set new-site's core DB masters (shards: s1-s8, x1, es4-es5) in read-write mode: 06-set-db-readwrite.py

Phase 7 - Set MediaWiki to read-write

1. Go to read-write mode by changing the ReadOnly conftool value: 07-set-readwrite.py

Phase 8 - post read-only

1. Start maintenance in the new DC: 08-start-maintenance.py
   1. Run puppet on the maintenance hosts
2. Update tendril for new database masters: 08-update-tendril.py Pure cosmetic change, no effect on production. No changes required for database zarcillo (which has a different master for eqiad and codfw).
3. Set the TTL for the DNS records to 300 seconds again: 08-restore-ttl.py
4. Update DNS records for new database masters deploying eqiad->codfw; codfw->eqiad This is not covered by the switchdc script. Please use the following to SAL log !log Phase 8.4 Update DNS records for new database masters
5. Make sure the CentralNotice banner informing users of readonly is removed. Keep in mind, there is some minor HTTP caching involved (~5mins)
6. Remove the downtime added in phase 0.

Phase 9 - verification and troubleshooting

This is not covered by the switchdc script

1. Make sure reading & editing works! :)
2. Make sure recent changes are flowing (see Special:RecentChanges, EventStreams, and the IRC feeds)
3. Make sure email works (exim4 -bp on mx1001/mx2001, test an email)

Dashboards

• Apache/HHVM
• App servers
• ATS cluster view (text)
• ATS backends<->Origin servers overview (appservers, api, restbase)
• MediaWiki errors
• Database Errors logs

Media storage/Swift

Switchover

Cache-wise, Swift is active-active. However the background synchronization swiftrepl needs to run in what Mediawiki considers the primary datacenter. A patch such as 622522 accomplishes this and needs to be merged after Mediawiki has been switched over (up to e.g. 24/48 hours later is acceptable)

Switching back

Revert 622522 to move swiftrepl back to eqiad

Dashboards

• Swift eqiad
• Swift codfw
• Thumbor
• ATS cluster view (upload)
• ATS backends<->Origin servers overview (swift)

ElasticSearch

General context on how to switchover

CirrusSearch talks by default to the local datacenter ($wmfDatacenter). If Mediawiki switches datacenter, elasticsearch will automatically follow.

Manually switching CirrusSearch to a specific datacenter can always be done. Point CirrusSearch to codfw by editing wmgCirrusSearchDefaultCluster InitialiseSettings.php.

To ensure coherence in case of lost updates, a reindex of the pages modified during the switch can be done by following Recovering from an Elasticsearch outage / interruption in updates.

Preserving more_like query cache performance

CirrusSearch has a caching layer that caches the result of Elasticsearch queries such as "more like this" queries (which are used, among other things, to generate "Related Articles" at the bottom of mobile Wikipedia pages).

Switching datacenters will result in degraded performance while the cache fills back up.

In order to avoid the aforementioned performance degradation, a mitigation should be deployed that will hardcode more_like queries to keep routing to the "old" datacenter for 24 hours following the switchover.

Hardcoding the cirrus cluster will allow the stampede of cache misses to be sent to the secondary search cluster which has enough capacity, once typical traffic has migrated to the new datacenter, to serve the increased load.

This hardcoding should be deployed in advance of the switchover. Since it is effectively a no-op until the actual cutover, it can be deployed as far in advance as desired.

For example, if we are switching over from eqiad to codfw, more_like queries should be hardcoded to route to eqiad; this change should be deployed before the actual cutover.

Then, 24 hours after the cutover, the hardcoding can be removed, allowing more_like queries to route to the new cirrus dc - in this example, codfw.

Days in advance preparation

Deploy a patch to hardcode more_like query routing to the currently active DC (i.e. the datacenter we are switching over from).

Example Patch: [mediawiki-config] 635411 cirrus: Temporarily hardcode more_like query routing

This mitigation should be left in place for 24 hours following the switchover (equivalent to the cache length), at which point there is no longer any performance penalty to removing the hardcoding.

One day after datacenter switch

Revert the earlier patch to hardcode more_like query routing; this will allow these queries to route to the newly active DC, and there will not be any performance degradation since the caches have been fully populated by this point.

Dashboards

• ElasticSearch Percentiles

Traffic

General information on generic procedures

https://wikitech.wikimedia.org/wiki/Global_Traffic_Routing

Switchover

GeoDNS (User-facing) Routing:

1. gerrit: C+2 and Submit commit https://gerrit.wikimedia.org/r/#/c/operations/dns/+/458806
2. <any authdns node>: authdns-update
3. SAL Log using the following !log Traffic: depool eqiad from user traffic

Switchback

Same procedure as above, with reversion of the commit specified: GeoDNS.

Dashboards

• Load Balancers
• Frontend traffic (text@eqiad/codfw)
• Frontend traffic (upload@eqiad/codfw)

Services

All services, are active-active in DNS discovery, apart from restbase, that needs special treatment. The procedure to fail over to one site only is the same for every one of them:

1. reduce the TTL of the DNS discovery records to 10 seconds
2. depool the datacenter we're moving away from in confctl / discovery
3. restore the original TTL

• All of the above is done using the sre.switchdc.services cookbooks:

# Switch the service "parsoid" to codfw-only
$ cookbook sre.switchdc.services --services parsoid -- eqiad codfw
# Switch all active-active services to codfw, excluding parsoid and cxserver
$ cookbook sre.switchdc.services --exclude parsoid cxserver -- eqiad codfw

Restbase is a bit of a special case, and needs an additional step, if we're just switching active traffic over and not simulating a complete failover:

1. pool restbase-async everywhere, then depool restbase-async in the newly active dc, so that async traffic is separated from real-users traffic as much as possible.

Dashboards

• EventStreams
• ORES
• Maps
• WDQS

Other miscellaneous

• Deployment server
• EventLogging
• IRC, RCStream, EventStreams

Schedule of past switches

Schedule for 2018 switch

You may have followed an incorrect link to this section. If this is the case, please see § Schedule for 2020 switch for current information.

• Services: Tuesday, September 11th 2018 14:30 UTC
• Media storage/Swift: Tuesday, September 11th 2018 15:00 UTC
• Traffic: Tuesday, September 11th 2018 19:00 UTC
• MediaWiki: Wednesday, September 12th 2018: 14:00 UTC

Switching back:

• Traffic: Wednesday, October 10th 2018 09:00 UTC
• MediaWiki: Wednesday, October 10th 2018: 14:00 UTC
• Services: Thursday, October 11th 2018 14:30 UTC
• Media storage/Swift: Thursday, October 11th 2018 15:00 UTC

Schedule for 2017 switch

See also "Editing pause for failover test" blog post (18 April 2017), on the Wikimedia Blog.

T138810 on Phabricator tracks tasks to be undertaken during the 2017 switch.

• Elasticsearch: elasticsearch is automatically following mediawiki switch
• Services: Tuesday, April 18th 2017 14:30 UTC
• Media storage/Swift: Tuesday, April 18th 2017 15:00 UTC
• Traffic: Tuesday, April 18th 2017 19:00 UTC
• MediaWiki: Wednesday, April 19th 2017 14:00 UTC (user visible, requires read-only mode)
• Deployment server: Wednesday, April 19th 2017 16:00 UTC

Switching back:

• Traffic: Pre-switchback in two phases: Mon May 1 and Tue May 2 (to avoid cold-cache issues Weds)
• MediaWiki: Wednesday, May 3rd 2017 14:00 UTC (user visible, requires read-only mode)
• Elasticsearch: elasticsearch is automatically following mediawiki switch
• Services: Thursday, May 4th 2017 14:30 UTC
• Swift: Thursday, May 4th 2017 15:30 UTC
• Deployment server: Thursday, May 4th 2017 16:00 UTC

Schedule for 2016 switch

See also "Wikimedia failover test" blog post (11 April 2016), on the Wikimedia Blog.

• Deployment server: Wednesday, January 20th 2016
• Traffic: Thursday, March 10th 2016
• MediaWiki 5-minute read-only test: Tuesday, March 15th 2016, 07:00 UTC
• Elasticsearch: Thursday, April 7th 2016, 12:00 UTC
• Media storage/Swift: Thursday, April 14th 2016, 17:00 UTC
• Services: Monday, April 18th 2016, 10:00 UTC
• MediaWiki: Tuesday, April 19th 2016, 14:00 UTC / 07:00 PDT / 16:00 CEST (requires read-only mode)

Switching back:

• MediaWiki: Thursday, April 21st 2016, 14:00 UTC / 07:00 PDT / 16:00 CEST (requires read-only mode)
• Services, Elasticsearch, Traffic, Swift, Deployment server: Thursday, April 21st 2016, after the above is done

Monitoring Dashboards

Aggregated list of interesting dashboards

• Apache/HHVM
• App servers
• Load Balancers
• Frontend traffic
• Varnish eqiad
• Varnish codfw
• Swift eqiad
• Swift codfw