Hi.
We are seeing problems when trying to upgrade from version 4.0.23 to
version 5.0.18 (using standard RPM's)
The upgrade includes new hardware and we are exporting our databases
with mysqldump and reading them in
on the new database server.
All seams to be just fine and enabling some traffic to the dbm is not a
problem, we have taken care of the old-passwords problem.
The tests rolled on for hours doing all types of queries, then problems
appeared.
The largest problem of them all as I see it is that we can not figure
out what is wrong.
Looking at the error log does not make me any smarter, but my hope is
that someone here might help me figure it out.
The manifestation of the problems seams to be that suddenly some
processes (queries) just start to hang, some doing temporary locks
hindering other queries from executing and putting them in cue. As the
cue increases we will soon have used up all our connections and
the whole dmb is unavailable. Those hung processes can not be killed and
will not go away over time.
As I said before I can not see anything in the error-log that can help
me understand this. All I see is InnoDB complaining about long semaphore
wait.
I don't know if and/or why this is the problem.
Snippet from log file:
---
060214 13:38:05 mysqld started
060214 13:38:05 [Warning] Changed limits: max_open_files: 65535
max_connections: 15000 table_cache: 25262
060214 13:38:05 InnoDB: Started; log sequence number 1 1237513011
060214 13:38:05 [Warning] Neither --relay-log nor --relay-log-index were
used; so replication may break when this MySQL server acts as a slave
and has his hostname changed!! Please use '--relay-log=saga-relay-bin'
to avoid this problem.
060214 13:38:05 [Note] /usr/sbin/mysqld: ready for connections.
Version: '5.0.18-standard-log' socket: '/var/lib/mysql/mysql.sock'
port: 3306 MySQL Community Edition - Standard (GPL)
InnoDB: Warning: a long semaphore wait:
--Thread 1394612576 has waited at btr0cur.c line 366 for 242.00 seconds
the semaphore:
S-lock on RW-latch at 0x2aaaaae5c120 created in file dict0dict.c line 3654
a writer (thread id 1255901536) has reserved it in mode exclusive
number of readers 0, waiters flag 1
Last time read locked in file btr0cur.c line 366
Last time write locked in file btr0cur.c line 359
InnoDB: Warning: a long semaphore wait:
--Thread 1368787296 has waited at btr0cur.c line 366 for 242.00 seconds
the semaphore:
S-lock on RW-latch at 0x2aaaaae5c120 created in file dict0dict.c line 3654
a writer (thread id 1255901536) has reserved it in mode exclusive
number of readers 0, waiters flag 1
Last time read locked in file btr0cur.c line 366
Last time write locked in file btr0cur.c line 359
---
The whole logfile can be found at:
https://saga.compricer.se/saga_publishlog.err
my.cnf:
---
[mysqld]max_connections = 15000
log-slow-queries = mysql_slow.log
long_query_time = 5
old-passwords
---
All possible help would be greatly appreciated
Regards
/Mattias
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
