Merge remote-tracking branch 'origin/master' into flake

[?]

Aug 9, 2019, 10:46 AM

WONFWQWSB35YEJ2KNZEXT4RHIQR2ZJ46IIJDHFM7PAASC6D3ZH5QC

Dependencies

[2] SW7STLQ7 Cache flake-based jobset evaluations
[3] TOLUNFAC Create extension pg_trgm in the NixOS module
[4] TM6C352U Ensure that stoptime and starttime are set for finished builds
[5] RJNMYHJF Add missing file
[6] QLOLZHRX Allow a per-jobset check interval
[7] GEADFVZ5 hydra-queue-runner: Improved scheduling
[8] W4G5MZZS hydra-evaluator improvements
[9] RXVJFQ5A Evaluator cleanups
[10] FANTYCR7 Materialize the number of finished builds
[11] KJQWSRCC Speed up hydra-update-gc-roots
[12] FTPCV25M Store aggregate members in the database
[13] BXHG3HYL When renaming a jobset, add a redirect from the old name
[14] D44B24QC Store the account type ("hydra" or "persona") explicitly in the database
[15] KN3VYE5P * Cleaned up the foreign key constraints.
[16] D5QIOJGP * Move everything up one directory.
[17] YU6CND7C Remove support for views
[18] RVHBV3B3 Notify the queue runner when a build is deleted
[19] IWXLV4OB Add DB columns for when to notify responsible committers and which inputs should be checked
[20] E7FID72S Remove the BuildMachines and BuildMachinesSystemTypes tables
[21] KSBB33RE Add a dashboard
[22] B7ENVLRS hydra-queue-runner: Make build notification more reliable
[23] G2T4WAHI Store the inputs of each evaluation in the database
[24] YTIDBFGU Drop unused "disabled" columns
[25] PCKLFRT5 Support push notification of repository changes
[26] APURLT2Q Revert "Apply IndexBuildsOnJobFinishedId to unfinished builds only"
[27] P57OYJ4P Index builds on stop time
[28] KGGQD6CO Fix the Darcs schema upgrade
[29] PHJF4FUT Add missing files
[30] DH3KNBAV Merge remote-tracking branch 'upstream/who-broke-builds' into upstream-master
[31] CHJANKMM IndexJobsetEvalsOnJobsetId: Only index jobsets with new builds
[32] 2JGYLIEC Speed up the listing of jobset evals by caching some info in the DB
[33] KDBWLKRP Index BuildSteps on propagatedFrom
[34] S2NXJOJE Fix a race that can cause hydra-queue-runner to ignore newly added builds
[35] H7HGWODE * Remove an empty file.
[36] 2NGV7NYN Drop unused "tag" column
[37] 26PYNDXV Remove the errorMsg column from the Builds table
[38] 4CELXP7P Remove the longDescription field
[39] LTQMSMYO Get rid of some obsolete indexes
[40] 4YCF3KBG Concurrent hydra-evaluator
[41] QUMWPGCU Add flake configuration to the web interface
[42] CCIORPJ5 When manually scheduling an eval, force re-instantiation of store derivations
[43] K2WAI5SO Apply IndexBuildsOnJobFinishedId to unfinished builds only
[44] 6COLDXKS Create a pg_trgm index on builds.drvpath
[45] RUTFLXFQ Drop unused BuildProducts.description column
[46] MOX7XJ2E Merge the BuildSchedulingInfo table into the Builds table
[47] TX7Q4RAS Add page showing latest build steps
[48] ACBS7C6Q hydra-queue-runner: Detect changes to the scheduling shares
[49] WE5Q2NVI Allow build to be bumped to the front of the queue via the web interface
[50] SS4TZXNU Distinguish between permanent evaluation errors and transient input errors
[51] G7KWXSFM Distinguish build step states
[52] FITVNQ2S Keep track of the time we spend copying to/from build machines
[53] BKMQXGBO Allow public dashboards
[54] EJ7C77C3 Allow searching for store and drv paths. Only does a complete match to reduce load on database.
[55] PLOZBRTR Add command ‘hydra-queue-runner --status’ to show current status
[56] T5BIOVJE Add support for tracking custom metrics
[57] E7M2WP7A Remove unused Jobs columns
[58] UJOSXBZI Record which build a failed build step was propagated from
[59] URHBNFTI Add forgotten file
[60] TPNHTE5V Remove obsolete Builds columns and provide accurate "Running builds"
[61] UNVMKJV5 Unify build and step status codes
[62] Q5HZWFCY Add support for darcs repositories.
[63] 53IMJNBB Add isChannel column and meta attribute.
[64] 2KLEQJTG sql: refactor some sql statements to lowercase
[65] FV2M6MOT hydra: use autoconf/-make
[66] W5OAZWPD Drop the errorMsg column in the Jobs table
[67] 6WRGCITD Enable declarative projects.
[68] JD7AWXQG Remove now-unused SystemTypes table
[69] EPWEMRI2 Allow determinism checking for entire jobsets
[70] 4HPT4SDD Revert "Remove now-unused SystemTypes table"
[71] 4S5JF5JP Use latest DBIx::Class::Schema::Loader
[72] YF3HTIPF Use partial indexes

Change contents

file deletion: upgrade-57.sql (----------)

[4.3004]→[4.321:359](∅→∅),[4.359]→[4.211:211](∅→∅)

-- The pg_trgm extension has to be created by a superuser. The NixOS
-- module creates this extension in the systemd prestart script. We
-- then ensure the extension has been created before creating the
-- index. If it is not possible to create the extension, a warning
-- message is emitted to inform the user the index creation is skipped
-- (slower complex queries on builds.drvpath).
do $$
begin
create extension if not exists pg_trgm;
-- Provide an index used by LIKE operator on builds.drvpath (search query)
create index IndexTrgmBuildsOnDrvpath on builds using gin (drvpath gin_trgm_ops);
exception when others then
raise warning 'Can not create extension pg_trgm: %', SQLERRM;
raise warning 'HINT: Temporary provide superuser role to your Hydra Postgresql user and run the script src/sql/upgrade-57.sql';
raise warning 'The pg_trgm index on builds.drvpath has been skipped (slower complex queries on builds.drvpath)';
end$$;
replacement in src/sql/upgrade-57.sql at line 1

[4.4169]→[4.4170:4556](∅→∅),[4.4556]→[2.1049:1096](∅→∅)

alter table Jobsets alter column nixExprInput drop not null;
alter table Jobsets alter column nixExprPath drop not null;
alter table Jobsets add column type integer default 0;
alter table Jobsets add column flake text;
alter table Jobsets add check ((type = 0) = (nixExprInput is not null and nixExprPath is not null));
alter table Jobsets add check ((type = 1) = (flake is not null));
alter table JobsetEvals add column flake text;

[4.4169]

-- The pg_trgm extension has to be created by a superuser. The NixOS
-- module creates this extension in the systemd prestart script. We
-- then ensure the extension has been created before creating the
-- index. If it is not possible to create the extension, a warning
-- message is emitted to inform the user the index creation is skipped
-- (slower complex queries on builds.drvpath).
do $$
begin
create extension if not exists pg_trgm;
-- Provide an index used by LIKE operator on builds.drvpath (search query)
create index IndexTrgmBuildsOnDrvpath on builds using gin (drvpath gin_trgm_ops);
exception when others then
raise warning 'Can not create extension pg_trgm: %', SQLERRM;
raise warning 'HINT: Temporary provide superuser role to your Hydra Postgresql user and run the script src/sql/upgrade-57.sql';
raise warning 'The pg_trgm index on builds.drvpath has been skipped (slower complex queries on builds.drvpath)';
end$$;
file addition: upgrade-58.sql (----------)

[4.3004]

alter table Jobsets alter column nixExprInput drop not null;
alter table Jobsets alter column nixExprPath drop not null;
alter table Jobsets add column type integer default 0;
alter table Jobsets add column flake text;
alter table Jobsets add check ((type = 0) = (nixExprInput is not null and nixExprPath is not null));
alter table Jobsets add check ((type = 1) = (flake is not null));
alter table JobsetEvals add column flake text;