Commit 45be5766 authored by Leo Gordon's avatar Leo Gordon
Browse files

added ChangeLog for the changes done during rel.69

parent 2cb1d998
......@@ -15,6 +15,559 @@ Summary:
---------------------[after EnsEMBL rel.69]----------------------------------------------------------------
2012-10-19 15:45 lg4
* sql/tables.sql: better match heavy queries with indices on job
2012-10-19 15:43 lg4
* modules/Bio/EnsEMBL/Hive/:, DBSQL/
merge reset_and_grab into one subroutine; pre-increment dependent
semaphore if re-running a DONE job; use -force flag for
force-running an individual job
2012-10-19 15:40 lg4
* scripts/ propagation of -force flag through
2012-10-17 12:55 lg4
* modules/Bio/EnsEMBL/Hive/:,,
DBSQL/ moved special-job-reset and
special-job-reclaim into the same call, removed the unnecessary
fetch in between
2012-10-16 12:37 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ cosmetic
2012-10-16 10:42 lg4
* modules/Bio/EnsEMBL/Hive/, scripts/ try not
to shock the Q::register_worker_death() code with inexistent
2012-10-16 10:26 lg4
* modules/Bio/EnsEMBL/Hive/, scripts/ moved
specializaton call into W::run, so that death messages during
specialization could be recorded in W->log_dir
2012-10-15 16:06 lg4
* modules/Bio/EnsEMBL/Hive/ print the resource_class_id
of the worker
2012-10-15 16:04 lg4
* scripts/ pass either rc_name or logic_name or job_id
from to
2012-10-15 10:44 mm14
* modules/Bio/EnsEMBL/Hive/DBSQL/ bugfix:
$analysis instead of $self->analysis
2012-10-15 10:42 lg4
* modules/Bio/EnsEMBL/Hive/ set compile_module_once=1 as
2012-10-13 12:31 lg4
* modules/Bio/EnsEMBL/Hive/:,
PipeConfig/ allow database names to contain
2012-10-13 11:02 lg4
* modules/Bio/EnsEMBL/Hive/, scripts/,
sql/patch_2012-10-13.sql, sql/tables.sql, sql/tables.sqlite: if is run manually, rc_name may stay NULL in the
2012-10-12 21:24 lg4
* docs/: hive_schema.mwb, hive_schema.png: updated schema diagram
with worker.resource_class_id
2012-10-12 17:15 lg4
* modules/Bio/EnsEMBL/Hive/,
modules/Bio/EnsEMBL/Hive/, scripts/,
sql/foreign_keys.mysql, sql/patch_2012-10-12.sql, sql/tables.sql,
sql/tables.sqlite: separating create_new_worker() from
2012-10-11 12:37 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/,
sql/triggers.mysql, sql/triggers.sqlite: proper counting of
semaphored jobs by triggers and in constructor
2012-10-10 14:45 lg4
* modules/Bio/EnsEMBL/Hive/, scripts/
refactoring of the Q::create_new_worker() and introduction of
-force flag
2012-10-10 14:36 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ we should
not leave SYNCHING analysis out (especially if there are not too
many READY analyses)
2012-10-10 14:34 lg4
* modules/Bio/EnsEMBL/Hive/RunnableDB/LongMult/
slow things down a little
2012-10-09 10:48 lg4
* docs/hive_schema.mwb, docs/hive_schema.png,
sql/foreign_keys.mysql: added a DF-to-DF foreign key and
refreshed the diagram
2012-10-09 10:25 lg4
* sql/tables.sqlite: bugfix: forgot to add semaphored_job_count to
SQLite schema, now included
2012-10-09 10:22 lg4
* sql/: patch_2012-10-08.sql, tables.sql, tables.sqlite: turned two
unique keys into primary keys (needed by BaseAdaptor)
2012-10-08 16:06 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ allow the
batch_size to be updated via
2012-10-08 12:17 lg4
* modules/Bio/EnsEMBL/Hive/, scripts/ removed
the input_id functionality from runWorker as both redundant
(standaloneJob) and probably not working
2012-10-08 12:13 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ those
"return" statements would have never worked anyway, so I removed
2012-10-05 16:14 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ extend the
param_init() of the garbage-collected jobs to include
analysis->parameters() for template substitution (still limited!)
2012-10-05 14:14 lg4
* modules/Bio/EnsEMBL/Hive/, sql/patch_2012-10-06.sql,
sql/tables.sql, sql/tables.sqlite: cause_of_death="" no longer
used for decision making, cause_of_death IS NULL by default and
FATALITY renamed UNKNOWN for clarity
2012-10-05 10:09 lg4
* modules/Bio/EnsEMBL/Hive/ fetch_failed_workers() is
dropped as no longer used, get_hive_current_load() cosmetically
2012-10-04 16:47 lg4
* modules/Bio/EnsEMBL/Hive/,
sql/patch_2012-10-05.sql, sql/tables.sql, sql/tables.sqlite:
EMPTY state added and definitions of READY and WORKING made more
2012-10-04 15:45 lg4
* modules/Bio/EnsEMBL/Hive/ bugfix: a typo
2012-10-04 15:39 lg4
* modules/Bio/EnsEMBL/Hive/,
scripts/, sql/patch_2012-10-04.sql, sql/tables.sql,
sql/tables.sqlite, sql/triggers.mysql, sql/triggers.sqlite:
introduced semaphored_job_count, renamed
unclaimed_job_count-->ready_job_count, changed reporting, fixed
2012-10-03 14:55 lg4
* modules/Bio/EnsEMBL/Hive/:,
DBSQL/ common denominator for
schedule_workers and specialize_new_worker
2012-10-03 14:11 lg4
* modules/Bio/EnsEMBL/Hive/:, DBSQL/
Fetching data via AnalysisAdaptor allows to print logic_names of
failed analyses
2012-10-03 11:09 lg4
* scripts/ print stats if could not create worker
anyway, but do not sync in the end (too cryptic)
2012-10-03 10:51 lg4
* modules/Bio/EnsEMBL/Hive/: DBSQL/,
PipeConfig/ renamed -input_job_id to
-prev_job_id to be in sync with other names
2012-10-02 16:47 lg4
* docs/: hive_schema.mwb, hive_schema.png: updated schema diagrams
2012-10-02 16:18 lg4
* modules/Bio/EnsEMBL/Hive/,
sql/patch_2012-10-02.sql, sql/tables.sql, sql/tables.sqlite:
moved failed_job_tolerance, max_retry_count, can_be_empty and
priority columns from analysis_stats to analysis_base
2012-10-02 14:56 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ bugfix: do
not forget PRE_- and POST_CLEANUP states
2012-10-02 13:00 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ bugfix:
changed the order of the atomic SEMAPHORED->READY state&counter
UPDATE so that it works as intended in SQLite as well
2012-10-02 12:17 lg4
* sql/tables.sql: added a fake default to last_update field
(required by stricter MySQL setup of Vega)
2012-10-02 11:48 lg4
* modules/Bio/EnsEMBL/Hive/,
sql/tables.sql, sql/tables.sqlite: added specific defaults into
analysis_stats and analysis_stats_monitor; re-ordered the fields
for easier navigation
2012-10-01 15:58 lg4
* modules/Bio/EnsEMBL/Hive/:, DBSQL/
bugfix: fetch_by_dbID should work now (thanks to ChuangKee and
2012-10-01 12:53 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ bugfix:
primary_key_constraint now works (thanks to Miguel!)
2012-09-28 11:01 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ bugfix:
typo fixed, thanks to Miguel for pointing out!
2012-09-27 16:48 lg4
* modules/Bio/EnsEMBL/Hive/:,
DBSQL/, PipeConfig/
make AnalysisStats a rearrangeable EnsEMBL-style constructor, a
proper store method and other preparations
2012-09-27 15:29 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/
simplification of DYNAMIC hive_capacity update code
2012-09-27 12:03 lg4
* scripts/ retired the script; likely not
working and duplicating functionality of more flexible PipeConfig
2012-09-27 10:50 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ bugfix:
also release jobs that were in PRE_CLEANUP or POST_CLEANUP states
2012-09-26 15:03 lg4
* modules/Bio/EnsEMBL/Hive/:, DBSQL/
switched Queen to become descendent of Hive::DBSQL::ObjectAdaptor
and removed _generic_fetch from it
2012-09-26 12:31 lg4
* modules/Bio/EnsEMBL/Hive/:, rearranged
Worker's storable getters/setters, introduced and used a proper
rearranging new() method
2012-09-26 11:27 lg4
* modules/Bio/EnsEMBL/Hive/:, Worker doesnt
really need its own reference to db (can go via adaptor)
2012-09-25 16:20 lg4
* modules/Bio/EnsEMBL/Hive/ bugfix: GROUP BY now includes
a proper prefix of the index
2012-09-25 16:04 lg4
* modules/Bio/EnsEMBL/Hive/,
sql/patch_2012-09-25.sql, sql/tables.sql, sql/tables.sqlite:
Dropped 'BLOCKED' job status and introduced 'SEMAPHORED' status
that is maintained in sync with semaphore_counts; less confusing
and more efficient (with new 3-part index)
2012-09-25 12:32 lg4
* modules/Bio/EnsEMBL/Hive/,
modules/Bio/EnsEMBL/Hive/, scripts/,
scripts/, sql/patch_2012-09-24.sql, sql/tables.sql,
sql/tables.sqlite: record each Workers log_dir in the database;
simplified the log_dir code and renamed cmdline options
2012-09-21 22:16 lg4
* docs/hive_schema.mwb, docs/hive_schema.png,
scripts/, sql/foreign_keys.mysql,
sql/patch_2012-09-21.sql, sql/tables.sql, sql/tables.sqlite:
moved resource_class_id from analysis_stats and
analysis_stats_monitor to analysis_base
2012-09-21 14:46 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/:, fetch_all never seems to be executed for
these adaptors
2012-09-21 09:34 lg4
* modules/Bio/EnsEMBL/Hive/Meadow/ bugfix: better parsing of
the LSF-job-name
2012-09-20 15:56 lg4
* modules/Bio/EnsEMBL/Hive/:,
DBSQL/ optimization: worker should not
sync analyses it is not ready to run
2012-09-20 11:51 lg4
* modules/Bio/EnsEMBL/Hive/,
modules/Bio/EnsEMBL/Hive/Meadow/, scripts/
replaced internal rc_id by rc_name in the Meadow code and in most
of the Scheduler; needs testing
2012-09-20 11:44 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ allow JOIN to
appear in the constraint and act wisely - so we do not need extra
complicated syntax for joining
2012-09-07 11:20 lg4
* modules/Bio/EnsEMBL/Hive/ not ready yet to scrap
the "Runnable" support
2012-09-07 10:29 lg4
* modules/Bio/EnsEMBL/Hive/ (patch offered by Matthieu)
Allow the Job to kill the Worker even on succecss
2012-09-05 15:07 mm14
* modules/Bio/EnsEMBL/Hive/RunnableDB/ Updated
the list of eHive tables
2012-09-05 15:00 lg4
* modules/Bio/EnsEMBL/Hive/:, DBSQL/,
DBSQL/ these methods are already defined
in the parent class
2012-09-05 10:33 lg4
* modules/Bio/EnsEMBL/Hive/:,
trimmed the commented-out 'use' statements
2012-09-04 17:07 lg4
* modules/Bio/EnsEMBL/Hive/ slimmed down the
Extensions module a bit (valuable code already moved into
2012-09-04 17:02 lg4
* docs/hive_schema.mwb, docs/hive_schema.png,
modules/Bio/EnsEMBL/Hive/Utils/, scripts/,
scripts/, sql/foreign_keys.mysql,
sql/patch_2012-09-04.sql, sql/procedures.mysql,
sql/procedures.sqlite, sql/tables.sql, sql/tables.sqlite:
substituted the overloaded legacy 'analysis' table by a slimmer
2012-09-04 10:09 lg4
* scripts/ actually switch to using rc_name in the
workers commandline
2012-09-03 12:26 lg4
* scripts/ make sure beekeeper reports the same
scheduling plans both when it is actually scheduling and in
"reporting" mode
2012-09-03 12:23 lg4
* scripts/ adding rc_name to the lsf_report
2012-09-03 12:21 lg4
* modules/Bio/EnsEMBL/Hive/Meadow/,
modules/Bio/EnsEMBL/Hive/Meadow/, scripts/
rc_name support in the beekeeper
2012-09-03 12:20 lg4
* modules/Bio/EnsEMBL/Hive/, scripts/ rc_name
support in the Worker
2012-08-29 09:51 lg4
* modules/Bio/EnsEMBL/Hive/ a typo in perldoc
2012-08-28 10:17 lg4
* modules/Bio/EnsEMBL/Hive/: DBSQL/, this
diagnostic information is no longer needed
2012-08-28 10:05 lg4
* modules/Bio/EnsEMBL/Hive/ Scheduler should explain that
workers are not added because of the pending ones
2012-08-27 14:10 mm14
* modules/Bio/EnsEMBL/Hive/RunnableDB/ eHive
tables are always included unless exclude_ehive is defined
2012-08-25 21:09 lg4
* modules/Bio/EnsEMBL/Hive/PipeConfig/ make
sure default is in the beginning of the list
2012-08-25 10:58 mm14
* modules/Bio/EnsEMBL/Hive/RunnableDB/ Updated
the eHive table list + fixed typo
2012-08-24 15:49 lg4
* modules/Bio/EnsEMBL/Hive/:,,,,, inherit
Job,Worker,DFR,RC from Bio::EnsEMBL::Storable, reuse some code
2012-08-24 15:44 lg4
* sql/tables.sql: starting the rel69...
2012-08-24 14:38 lg4
* modules/Bio/EnsEMBL/Hive/ bugfix: make sure there is at
least a number in the query (reported by Matthieu)
2012-08-23 12:01 lg4
* modules/Bio/EnsEMBL/Hive/:, renamed to_string into toString for
2012-08-23 10:45 mm14
* modules/Bio/EnsEMBL/Hive/RunnableDB/ Added a
"skip_dump" parameter to ease the restoration of a dump
2012-08-17 15:52 lg4
* modules/Bio/EnsEMBL/Hive/ trying to be more careful
with references; release jobs' parameters earlier
2012-08-16 12:16 mm14
* modules/Bio/EnsEMBL/Hive/RunnableDB/ Can copy a
database to another database
2012-08-16 12:12 lg4
* modules/Bio/EnsEMBL/Hive/:, change
suggested by Matthieu to avoid crashing if the temp_directory has
already been deleted by Runnable
2012-08-14 11:57 lg4
* modules/Bio/EnsEMBL/Hive/,
scripts/, scripts/,
scripts/, scripts/ switch to
module->new() notation everywhere, to simplify text searches
2012-08-03 16:31 lg4
* scripts/ no longer used as individual jobs are
no longer specifically blocked
2012-08-03 16:22 lg4
* modules/Bio/EnsEMBL/Hive/ removed honeycomb support
because it is no longer used by Compara modules
2012-08-03 10:36 lg4
* modules/Bio/EnsEMBL/Hive/DBSQL/ hopefully
will fix the "was not locked with LOCK TABLES" error message
---------------------[before EnsEMBL rel.69]----------------------------------------------------------------
2012-08-01 14:23 lg4
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment