Hi... i'm getting some many errors like this... Mon Aug 20 11:27:26 2012 : Error: [sql] Couldn't update SQL accounting STOP record - Lock wait timeout exceeded; try restarting transaction Mon Aug 20 11:27:26 2012 : Error: rlm_sql_mysql: Cannot store result Mon Aug 20 11:27:26 2012 : Error: rlm_sql_mysql: MySQL error 'Lock wait timeout exceeded; try restarting transaction' and other lines with Logins OK... but have more errors sql than logins ok... i guess missing some tunning parameters... i have, 15 NAS vs 2 freeradius-servers ..... this do AA vs 2 LDAP balanced and A vs cluster-mysql can help me more less to up or down some numbers to tune efficient the requests about some experience ? thx any way -- Antonio Peña Secure email with PGP 0x8B021001 available at http://pgp.mit.edu Fingerprint: 74E6 2974 B090 366D CE71 7BB2 6476 FA09 8B02 1001
Hi,
i'm getting some many errors like this... Mon Aug 20 11:27:26 2012 : Error: [sql] Couldn't update SQL accounting STOP record - Lock wait timeout exceeded; try restarting transaction Mon Aug 20 11:27:26 2012 : Error: rlm_sql_mysql: Cannot store result Mon Aug 20 11:27:26 2012 : Error: rlm_sql_mysql: MySQL error 'Lock wait timeout exceeded; try restarting transaction' and other lines with Logins OK... but have more errors sql than logins ok... i guess missing some tunning parameters... i have, 15 NAS vs 2 freeradius-servers ..... this do AA vs 2 LDAP balanced and A vs cluster-mysql can help me more less to up or down some numbers to tune efficient the requests about some experience ? thx any way
first thing I would suggest is to stop doing LIVE inserts of your accounting - eg use buffered_sql module instead second thing - change your DB engine to InnoDB instead of MyISAM thurd thing - add indexes and start a large number of DB connectors. alan
On 20 Aug 2012, at 16:40, alan buxey <A.L.M.Buxey@LBORO.AC.UK> wrote:
Hi,
i'm getting some many errors like this... Mon Aug 20 11:27:26 2012 : Error: [sql] Couldn't update SQL accounting STOP record - Lock wait timeout exceeded; try restarting transaction Mon Aug 20 11:27:26 2012 : Error: rlm_sql_mysql: Cannot store result Mon Aug 20 11:27:26 2012 : Error: rlm_sql_mysql: MySQL error 'Lock wait timeout exceeded; try restarting transaction' and other lines with Logins OK... but have more errors sql than logins ok... i guess missing some tunning parameters... i have, 15 NAS vs 2 freeradius-servers ..... this do AA vs 2 LDAP balanced and A vs cluster-mysql can help me more less to up or down some numbers to tune efficient the requests about some experience ? thx any way
first thing I would suggest is to stop doing LIVE inserts of your accounting - eg use buffered_sql module instead
second thing - change your DB engine to InnoDB instead of MyISAM
It's NDB (he's running cluster). -Arran
hi yes I have 64 to start initial each radius-server and yes have NDB engine into cluster.. .just missing the use for buffered_sql for live session.. that i have to read to implement, for example now.. not happening.. just when occurs start to doing that issues.. thanxs for the tips 2012/8/20 Arran Cudbard-Bell <a.cudbardb@freeradius.org>
On 20 Aug 2012, at 16:59, alan buxey <A.L.M.Buxey@lboro.ac.uk> wrote:
Hi,
It's NDB (he's running cluster).
ah yes. i skimmed over that bit - its obvious then. move to postgresSQL ;-)
and hack in his own fully redundant clustering system? ;)
-Arran
- List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html
-- Antonio Peña Secure email with PGP 0x8B021001 available at http://pgp.mit.edu Fingerprint: 74E6 2974 B090 366D CE71 7BB2 6476 FA09 8B02 1001
On 20 Aug 2012, at 16:33, Tony Peña <emperor.cu@gmail.com> wrote:
Hi...
i'm getting some many errors like this...
Mon Aug 20 11:27:26 2012 : Error: [sql] Couldn't update SQL accounting STOP record - Lock wait timeout exceeded; try restarting transaction Mon Aug 20 11:27:26 2012 : Error: rlm_sql_mysql: Cannot store result Mon Aug 20 11:27:26 2012 : Error: rlm_sql_mysql: MySQL error 'Lock wait timeout exceeded; try restarting transaction'
and other lines with Logins OK... but have more errors sql than logins ok...
i guess missing some tunning parameters... i have, 15 NAS vs 2 freeradius-servers ..... this do AA vs 2 LDAP balanced and A vs cluster-mysql can help me more less to up or down some numbers to tune efficient the requests about some experience ?
Are the queries wrapped in the transaction taking a long time to complete? Have you enabled the slow query log? Is there a firewall in between the RADIUS servers and the cluster that could be ungracefuly terminating your TCP connections? -Arran
participants (3)
-
alan buxey -
Arran Cudbard-Bell -
Tony Peña