Closed Bug 641067 Opened 14 years ago Closed 13 years ago

update sync sql schema to match production

Categories

(Cloud Services Graveyard :: Server: Sync, defect)

x86
macOS
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: Atoll, Assigned: telliott)

Details

(Whiteboard: [php][py=641068])

we discovered that the production schema does not match what is in hg, leading to data issues in sync1..5. please verify that the sync sql data in hg matches production. CREATE TABLE `wbo22` ( `username` int(11) NOT NULL, `collection` smallint(6) NOT NULL DEFAULT '0', `id` varbinary(64) NOT NULL DEFAULT '', `parentid` varbinary(64) DEFAULT NULL, `predecessorid` varbinary(64) DEFAULT NULL, `sortindex` int(11) DEFAULT NULL, `modified` bigint(20) DEFAULT NULL, `payload` longtext, `payload_size` int(11) DEFAULT NULL, `ttl` int(11) DEFAULT NULL, PRIMARY KEY (`username`,`collection`,`id`), KEY `ttl_idx` (`ttl`) ) ENGINE=InnoDB DEFAULT CHARSET=latin1
Whiteboard: [php] → [php][py=641068]
Depends on: 641086
No longer depends on: 641086
Isn't there a KEY `modified` (`username`,`collection`,`modified`), in there? That should be essential for all the "select * from wbo where uid = foo and collection = bar and modified > X" queries that the server does. We did decide to get rid of the payload_size index and just take the hit when we aren't in memcache, but this one seems crucial.
That seems reasonable to me, but it's not present on the samples I checked, sorry.
Ugh. That means we've had a schema in prod that has never matched anything we've tested. OTOH, it seems to be holding up OK, which leaves me with a dilemma.
Advise committing the prod schema as written, and then opening a bug to fix it to match what you think it should be.
no more php
Status: NEW → RESOLVED
Closed: 13 years ago
Resolution: --- → FIXED
Product: Cloud Services → Cloud Services Graveyard
You need to log in before you can comment on or make changes to this bug.