This is sort of related to my last post about slow queries due to
sorting, but I've noticed that the SIGTERM doesn't (quickly) abort any
DB requests, even if its a select (I can understand not interrupting
anything transactional)
ie from a test run
015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS |
==================== Dumping stats at 2015/04/05 21:03:17
==============
=======
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | ========
General statistics =========
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | Daemon
start time: 2015/04/05 13:03:17
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | Started
modules: purge
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | =======
Resource Monitor stats ======
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | Trigger #0
(periodic) : purge running.
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS |
============ Purge stats ============
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | idle purge
threads = 8
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | purge
operations pending = 0
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | purge status:
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS |
successfully purged = 500000
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | total
purged volume = 97266528256 (90.59 GB)
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | last file
submitted 901 s ago
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | last file
handled 901 s ago
2015/04/05 21:03:17 robinhood@magnus-data1[12418/3] STATS | last file
purged 901 s ago
2015/04/05 21:03:40 robinhood@magnus-data1[12418/2] SigHdlr | SIGTERM
received: performing clean daemon shutdown
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS |
==================== Dumping stats at 2015/04/05 21:23:17
==============
=======
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | ========
General statistics =========
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | Daemon
start time: 2015/04/05 13:03:17
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | Started
modules: purge
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | =======
Resource Monitor stats ======
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | Trigger #0
(periodic) : purge running.
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS |
============ Purge stats ============
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | idle purge
threads = 8
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | purge
operations pending = 0
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | purge status:
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS |
successfully purged = 500000
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | total
purged volume = 97266528256 (90.59 GB)
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | last file
submitted 2101 s ago
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | last file
handled 2101 s ago
2015/04/05 21:23:17 robinhood@magnus-data1[12418/3] STATS | last file
purged 2101 s ago
you can see the SIGTERM go in at
2015/04/05 21:03:40 robinhood@magnus-data1[12418/2] SigHdlr | SIGTERM
received: performing clean daemon shutdown
but looking on the DB server it's still building that query
MariaDB [(none)]> show processlist ;
+-----+-----------+------------------------------------+-------------+---------+------+----------------+------------------------------------------------------------------------------------------------------+----------+
| Id | User | Host | db |
Command | Time | State | Info
|
Progress |
+-----+-----------+------------------------------------+-------------+---------+------+----------------+------------------------------------------------------------------------------------------------------+----------+
| 278 | robinhood | magnus-data1.pawsey.ivec.org:37893 | rbh_scratch |
Sleep | 3193 | | NULL
|
0.000 |
| 279 | robinhood | magnus-data1.pawsey.ivec.org:37895 | rbh_scratch |
Sleep | 3193 | | NULL
|
0.000 |
| 282 | robinhood | magnus-data1.pawsey.ivec.org:37898 | rbh_scratch |
Sleep | 3193 | | NULL
|
0.000 |
| 281 | robinhood | magnus-data1.pawsey.ivec.org:37897 | rbh_scratch |
Sleep | 3193 | | NULL
|
0.000 |
| 283 | robinhood | magnus-data1.pawsey.ivec.org:37900 | rbh_scratch |
Sleep | 3193 | | NULL
|
0.000 |
| 284 | robinhood | magnus-data1.pawsey.ivec.org:37899 | rbh_scratch |
Query | 3193 | Sorting result | SELECT id FROM ENTRIES WHERE
(ENTRIES.invalid=0 OR ENTRIES.invalid IS NULL) AND ENTRIES.type<>'dir'
| 0.000 |
| 285 | robinhood | magnus-data1.pawsey.ivec.org:37902 | rbh_scratch |
Sleep | 3193 | | NULL
|
0.000 |
| 286 | robinhood | magnus-data1.pawsey.ivec.org:37903 | rbh_scratch |
Sleep | 3193 | | NULL
|
0.000 |
| 287 | robinhood | magnus-data1.pawsey.ivec.org:37904 | rbh_scratch |
Sleep | 3193 | | NULL
|
0.000 |
| 410 | root | localhost | NULL |
Query | 0 | init | show processlist
|
0.000 |
+-----+-----------+------------------------------------+-------------+---------+------+----------------+------------------------------------------------------------------------------------------------------+----------+
10 rows in set (0.00 sec)
(I know that I need to find a better DB host, but figured this
probably affects all users, even if they don't notice it as badly
and finally...
2015/04/05 21:43:13 robinhood@magnus-data1[12418/4] Purge | Purge
aborted, stop enqueuing purge requests.
2015/04/05 21:43:13 robinhood@magnus-data1[12418/4] ResMonitor | Purge
aborted after releasing 500000 entries, 189973688 blocks in /
scratch/director1080/2014.
2015/04/05 21:43:13 robinhood@magnus-data1[12418/4] ResMonitor |
Trigger #0 check function returned error 125... Will retry later
2015/04/05 21:43:13 robinhood@magnus-data1[12418/4] ResMonitor |
Current usage max is 0.00%
2015/04/05 21:43:13 robinhood@magnus-data1[12418/1] Main |
ResourceMonitor terminated its task
2015/04/05 21:43:13 robinhood@magnus-data1[12418/1] Main | All tasks
done! Exiting.
Ah, the joys of openfoam users (hence why I'm using this part of the
tree for scalng tests)
magnus-data1:/var/log # rbh-report -i
type , count, volume, avg_size
dir , 17329091, 66.50 GB, 4.02 KB
file , 68862002, 10.14 TB, 158.07 KB
Total: 86191093 entries, 11217891181084 bytes (10.20 TB)
Andrew
------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
robinhood-support mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/robinhood-support
