summaryrefslogtreecommitdiff
path: root/tests
AgeCommit message (Collapse)Author
2013-03-22tests/londiste/regen.sh: test various combination of optionsMarko Kreen
2013-03-01create-node: validate, accept public location from configMarko Kreen
2012-11-29tests/londiste: testcase for weird mergeMarko Kreen
2012-11-29remove obsolete filesMarko Kreen
2012-11-28Few more syncer fixesMarko Kreen
2012-11-28More tests for londisteMarko Kreen
2012-11-28londiste regen.sh: don't run repair/compare quietlyMarko Kreen
2012-11-28testlib.sh: use CASCADE when dropping tableMarko Kreen
2012-11-26londiste resurrect: Get dead root back into cascadeMarko Kreen
In some cases the 'takeover --dead-root' will be used although actual data on node is fine. Eg. network downtime, lost power. It would be good to have a way to bring it back to cascade without need for full rebuild. This command sync queue contents and re-subscribes node. Any events found on old root that are not spread to rest of cascade are dumped and deleted. No actual table changes are made, thus there must be external mechanism to avoid data conflicts.
2012-10-18Two merge related tests created.Asko Oja
merge_all tests simple merge from two partitions into one target Also tests adding columns through cascade merge_qnode tests same thing as merge all but uses also qnodes
2012-10-08New syncer logic for non-root providerMarko Kreen
Table-based locking is broken if provider is not root as it blocks whole node. Implement different logic for such case that pauses upstream worker to get well-defined sync point.
2012-07-18tests/merge/addcol-data2.sql: missing fileMarko Kreen
2012-07-18tests/simplecons - simple_consumer & simple_local_consumer testsMarko Kreen
2012-07-13londiste: --find-copy-node switch for add-tableMarko Kreen
This will find node where tables are available for copy by walking upwards to root.
2012-07-13londiste execute: fix downstream applyMarko Kreen
- Downgrade "already applied" message from WARN to INFO Unnecessary to warn about expected situation. - ExecAttr: urldecoding fixes This path was not tested properly. - Test execute-through-combined-queue
2012-07-13pgq.LocalConsumer: consumer that tracks processed ticks in local fileMarko Kreen
2012-07-13testlib.sh: few improvementsMarko Kreen
title2, title3, run_qadmin, cat_file
2012-07-05londiste: Wait commands for londisteMarko Kreen
add-table --wait-sync Don't return until all pending tables are synced wait-sync Wait until all pending tables are synced. wait-root Wait until root's next tick has appeared locally. wait-provider Wait until provider's next tick has appeared locally.
2012-01-31--sync-watermark: limit watermark to specific nodesMarko Kreen
This will allow specified nodes to sync global watermark with each other, without sending it upwards. This isolates root node from lag on downstream nodes.
2011-12-29tests/handler: test dispatcher tooMarko Kreen
2011-12-23tests/merge: make test run againMarko Kreen
Now local_add adds table to combined-root too. Test did not consider that.
2011-12-14tests/londiste: test repairMarko Kreen
2011-11-21tests: replace 'replay' with 'worker'Marko Kreen
2011-11-01tests/rename: test for --dest-tableMarko Kreen
2011-10-24tests/londiste: sanitize a bitMarko Kreen
turn off log grep to make output human-viewable
2011-10-24testlib: use code blocksMarko Kreen
2011-10-04tests/londiste: test pgq_autocommitMarko Kreen
2011-08-22tests/noqueue: minor improvementsMarko Kreen
- avoid hardcoded sleeps - avoid hardcoded part count
2011-08-07multimaster handlerEgon Valdmees
2011-06-15part handlerEgon Valdmees
handler adds trigger argument to full extra3 field with hash value and filter to event_processing and full_copy
2011-06-08fixed merge testcodeEgon Valdmees
2011-06-07fake_local and qsplitter handlersEgon Valdmees
added new handlers: * fake_local - dummy handler to setup queue tables. All events are ignored. Table structure is not required. Use in branch/leaf * qsplitter - dummy handler to setup queue tables. All events are ignored. Table structure is not required. All table events are inserted to destination queue, specified with handler arg 'queue'. changed local_add_table to allow adding of non-existing tables
2011-05-11OCM-2279: londiste3 should translate everything to utf8, in a lossy wayEgon Valdmees
where appropriate * added argument 'encoding' to dispatcher handler * tests for invalid utf8 sequences * support for renamed table copy in dispatcher handler
2011-04-15OCM-1524: merge_all switch for non-queue merge leafEgon Valdmees
2011-04-15OCM-1524: syntax error fix and added testsEgon Valdmees
2011-04-15OCM-1524: adding table to n:1 cascade merge should work as with merge-leaf ↵Egon Valdmees
scenario
2011-01-24pgq/cascade/admin: make tag-dead store into dbMarko Kreen
2011-01-24londiste: change behaviour of --create* flagsMarko Kreen
--create now does minimal pkey-only structure --create-full does maximal structure --create-only is dropped as too complex
2011-01-12tests/londiste: test lagging takeoverMarko Kreen
2011-01-10tests: default to tcp connectionMarko Kreen
2011-01-10tests/londiste: generate some load when testingMarko Kreen
2010-12-03tests/handler: bulk handler testMarko Kreen
2010-12-01tests/londiste: test table-remove on subscribersMarko Kreen
2010-11-29tests/merge: make sync fasterMarko Kreen
2010-11-29tests/merge: wait for sync on combined-rootMarko Kreen
2010-11-25tests/merge: some improvementsMarko Kreen
2010-11-24tests/merge: make it work againMarko Kreen
2010-11-24testlib: nicer psql launcherMarko Kreen
2010-11-19testlib cleanup fixesMarko Kreen
2010-06-25Update londiste test with new script namesMarko Kreen