Eqiad Migration Planning/Checklist
From Wikitech
< Eqiad Migration Planning(Difference between revisions)
| Line 6: | Line 6: | ||
* Ceph at EQIAD | * Ceph at EQIAD | ||
| + | ** can we serve images | ||
| + | ** does upload wizard function | ||
* Image scalers | * Image scalers | ||
| + | ** this will be shown by upload wizard functionality | ||
* Apache servers | * Apache servers | ||
| + | ** does anything show up :) | ||
* git-deploy | * git-deploy | ||
| + | ** is prereq of whole cutover | ||
* cluster replication | * cluster replication | ||
* Math, Captcha, Misc objects from ms7 to Swift | * Math, Captcha, Misc objects from ms7 to Swift | ||
* memcached | * memcached | ||
* redis | * redis | ||
| + | ** this will be tested by sessions. | ||
| + | ** might also want to do a couple of manual tests | ||
* parser cache, parser cache sharding | * parser cache, parser cache sharding | ||
| + | ** replication already functioning, but need to tail error log | ||
* database servers replication | * database servers replication | ||
| + | ** heartbeat will tell us the status of this | ||
* proper SQL grants in place | * proper SQL grants in place | ||
| + | ** done by asher | ||
* poolcounter | * poolcounter | ||
* Netapp discrepancies handled | * Netapp discrepancies handled | ||
* logging: wmerrors + apache syslog | * logging: wmerrors + apache syslog | ||
| + | ** this can be pre-tested by taking a look on fluorine | ||
* parsoid servers | * parsoid servers | ||
* Varnish | * Varnish | ||
Revision as of 22:08, 14 January 2013
EQIAD Checklist
Provision:
- Ceph at EQIAD
- can we serve images
- does upload wizard function
- Image scalers
- this will be shown by upload wizard functionality
- Apache servers
- does anything show up :)
- git-deploy
- is prereq of whole cutover
- cluster replication
- Math, Captcha, Misc objects from ms7 to Swift
- memcached
- redis
- this will be tested by sessions.
- might also want to do a couple of manual tests
- parser cache, parser cache sharding
- replication already functioning, but need to tail error log
- database servers replication
- heartbeat will tell us the status of this
- proper SQL grants in place
- done by asher
- poolcounter
- Netapp discrepancies handled
- logging: wmerrors + apache syslog
- this can be pre-tested by taking a look on fluorine
- parsoid servers
- Varnish
- Mediawiki config in place
- failover (both directions)
- test/test2 or equivalent supported (optionally name test -> pre-production and test2 -> staging to reflect actual function)
Tests:
- Ceph objects accessed correctly
- Apaches replicated properly (Jeff Green replay script to test this)
- git-deploy dry runs
- logging
- failover/fail back
UI tests:
- Login works
- Logout works
- UploadWizard works
- Upload new file over existing image works
- Image purge works
- Anonymous editing works
- Math extension works
- PDF book creation works
- Video transcoding works
- Job queue is processing
- Captcha (account creation)
- Make edits on wikis associated with each database cluster (s1, s2...s7)
- User
- registers
- search article
- read article
- comment on article
- edit article
- create article
- localization
- Community member
- tag article
- (exercise special pages features)
- exercise special page functions
