external
/
citus
mirror of https://github.com/citusdata/citus.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
5,398 Commits
645 Branches
162 Tags
103 MiB
Commit Graph

5398 Commits (7ee4646e9693754aa7ebbc654f6a6003f1c5f793)

Author SHA1 Message Date
Nils Dijk 7ee4646e96
use patternmatcher for join pushdown 2021-12-01 12:54:42 +01:00
Nils Dijk f4c5f9e2ea
remove liblwgeom from dependencies in citus 2021-12-01 12:54:41 +01:00
Nils Dijk 089650db11
rewriting tree to distributed plan 2021-12-01 12:54:41 +01:00
Nils Dijk 1aa86f6b1c
more simplification and removal of unused code 2021-12-01 12:54:41 +01:00
Nils Dijk 4362502bb5
generic way of matching fields for Var/Const 2021-12-01 12:54:41 +01:00
Nils Dijk 2b4072ebe8
rework the capture logic 2021-12-01 12:54:41 +01:00
Nils Dijk 766ed2ad33
collapse closing braces as suggested by dim 2021-12-01 12:54:40 +01:00
Nils Dijk cc84b48fd1
formatting and cleanup 2021-12-01 12:54:40 +01:00
Nils Dijk e3953b2459
prototype of pattern matcher 2021-12-01 12:54:40 +01:00
Nils Dijk bac16a3785
basic geo partitioning planning 2021-12-01 12:54:40 +01:00
Nils Dijk ec994a3a9c
add sql script to initialize a geo distributed table 2021-12-01 12:54:40 +01:00
Nils Dijk ff187ce5e1
expand shard model for geo distribution 2021-12-01 12:54:40 +01:00
Nils Dijk 6bc9209baf
update pg and citus 2021-12-01 12:54:39 +01:00
Nils Dijk b5ac6d2048
add broadcast optimization rule 2021-12-01 12:54:39 +01:00
Nils Dijk f952d9c614
easily add new join optimization functions 2021-12-01 12:54:39 +01:00
Nils Dijk 6b8e0a6105
fix subtle bug in adding to a list we are iterating over 2021-12-01 12:54:39 +01:00
Nils Dijk 23b11b470b
add paths for joins instead of rewriting them 2021-12-01 12:54:39 +01:00
Nils Dijk 84068f79b2
fix refactor and renames 2021-12-01 12:54:38 +01:00
Nils Dijk 2e178d21d6
fix targetlist to prevent projection that failed 2021-12-01 12:54:38 +01:00
Nils Dijk 5cef5b20a3
simple group by push down 2021-12-01 12:54:38 +01:00
Nils Dijk e570149c0a
propagate partition value during joins 2021-12-01 12:54:38 +01:00
Nils Dijk 30d9380f46
implement beginning of shard pruning 2021-12-01 12:54:38 +01:00
Nils Dijk ffa33b096a
Add support for optimising nested loop a bit (primary key lookup). 2021-12-01 12:54:38 +01:00
Nils Dijk 2937d4edc9
Fix misread of colocation id 2021-12-01 12:54:37 +01:00
Nils Dijk f858efd2f2
Implement beginnings of join pushdown 2021-12-01 12:54:37 +01:00
Nils Dijk b361386e05
Record the colocation id on the DU 2021-12-01 12:54:37 +01:00
Nils Dijk 8d73ff6779
Define a custom path structure 2021-12-01 12:54:37 +01:00
Nils Dijk 9acf41a9d9
Remove RTE reference from DU 2021-12-01 12:54:37 +01:00
Nils Dijk 00083afeaa
Use path object to recreate range table entry etc 2021-12-01 12:54:36 +01:00
Nils Dijk 285f1b1d1e
move path based planner to its own files 2021-12-01 12:54:36 +01:00
Nils Dijk fc0e6a1094
hacky hack for double distributed join 2021-12-01 12:54:36 +01:00
Nils Dijk aa3f283d02
support multiple remote scans 2021-12-01 12:54:36 +01:00
Nils Dijk 0d55755951
recreate query struct for sub paths 2021-12-01 12:54:36 +01:00
Nils Dijk 8f7037da63
beginning of set_rel_pathlist_hook to plan query on distributed table 2021-12-01 12:54:34 +01:00
Önder Kalacı ab365a335d
Merge pull request #5486 from citusdata/disable_node_async 
Allow disabling node(s) when multiple failures happen
 2021-12-01 10:48:49 +01:00
Onder Kalaci 549edcabb6 Allow disabling node(s) when multiple failures happen 
As of master branch, Citus does all the modifications to replicated tables
(e.g., reference tables and distributed tables with replication factor > 1),
via 2PC and avoids any shardstate=3. As a side-effect of those changes,
handling node failures for replicated tables change.

With this PR, when one (or multiple) node failures happen, the users would
see query errors on modifications. If the problem is intermitant, that's OK,
once the node failure(s) recover by themselves, the modification queries would
succeed. If the node failure(s) are permenant, the users should call
`SELECT citus_disable_node(...)` to disable the node. As soon as the node is
disabled, modification would start to succeed. However, now the old node gets
behind. It means that, when the node is up again, the placements should be
re-created on the node. First, use `SELECT citus_activate_node()`. Then, use
`SELECT replicate_table_shards(...)` to replicate the missing placements on
the re-activated node.
 2021-12-01 10:19:48 +01:00
Halil Ozan Akgül 6feb009834
Merge pull request #5499 from citusdata/fix_enterprise_fails 
Fix enterprise fails
 2021-11-30 16:16:30 +03:00
Halil Ozan Akgul 316274b5f0 Add normalize.sed item for multi_fix_partition_shard_index_names test 2021-11-30 13:28:41 +03:00
Halil Ozan Akgul 11072b4cb8 Normalize create role command in drop_partitioned_table test 2021-11-30 12:46:22 +03:00
Onur Tirtir 1836361a51
Add changelog entries for 10.2.3 (#5498) 2021-11-29 11:48:00 +03:00
Önder Kalacı 5ef0bae06f
Merge pull request #5493 from citusdata/metadata_connecttion 
Make sure to use a dedicated metadata connection
 2021-11-26 14:47:49 +01:00
Onder Kalaci d405993b57 Make sure to use a dedicated metadata connection 
With this commit, we make sure to use a dedicated connection per
node for all the metadata operations within the same transaction.

This is needed because the same metadata (e.g., metadata includes
the distributed table on the workers) can be modified accross
multiple connections.

With this connection we guarantee that there is a single metadata connection.
But note that this connection can be used for any other operation.
In other words, this connection is not only reserved for metadata
operations.
 2021-11-26 14:36:28 +01:00
Önder Kalacı 7b6588fec0
Merge pull request #5469 from citusdata/make_errors_generic 
Generalize the error checks while removing node
 2021-11-26 14:31:26 +01:00
Onder Kalaci 38b08ebde9 Generalize the error checks while removing node 
The checks for preventing to remove a node are very much reference
table centric. We are soon going to add the same checks for replicated
tables. So, make the checks generic such that:
	 (a) replicated tables fit naturally
	 (b) we can the same checks in `citus_disable_node`.
 2021-11-26 14:25:29 +01:00
Hanefi Onaldi 4c135de9e4 Introduce CI checks for hash comments in specs 
We do not use comments starting with # in spec files because it creates
errors from C preprocessor that expects directives after this character.
Instead use C style comments, i.e:
// single line comment

You can also use multiline comments as well
/*
 * multi line comment
 */
 2021-11-26 14:52:51 +03:00
Halil Ozan Akgül 27a161f443
Merge pull request #5496 from citusdata/multi-1-schedule-with-metadata-syncing 
Fix tests in multi-1-schedule that fail with metadata syncing
 2021-11-26 13:28:18 +03:00
Halil Ozan Akgul 87a1c760d9 Fix tests in multi-1-schedule that fail with metadata syncing 2021-11-26 12:09:53 +03:00
Önder Kalacı 0beb1aba62
Merge pull request #5470 from citusdata/redefine_active_placements 
Active placements can only be on active nodes
 2021-11-26 09:23:28 +01:00
Onder Kalaci 121f5c4271 Active placements can only be on active nodes 
We re-define the meaning of active shard placement. It used
to only be defined via shardstate == SHARD_STATE_ACTIVE.

Now, we also add one more check. The worker node that the
placement is on should be active as well.

This is a preparation for supporting citus_disable_node()
for MX with multiple failures at the same time.

With this change, the maintanince daemon only needs to
sync the "node metadata" (e.g., pg_dist_node), not the
shard metadata.
 2021-11-26 09:14:33 +01:00
Önder Kalacı d6cbfd0886
Merge pull request #5467 from citusdata/remove_useless_locks 
Do not acquire locks on reference tables when a node is removed/disabled
 2021-11-26 09:13:55 +01:00