Mitch, ----- Alkuperäinen viesti ----- Lähettäjä: "Mitch Pirtle" <[EMAIL PROTECTED]> Vastaanottaja: "Heikki Tuuri" <[EMAIL PROTECTED]> Kopio: <[EMAIL PROTECTED]> Lähetetty: Monday, July 05, 2004 4:26 PM Aihe: Re: InnoDB and long semaphore waits
> Heikki Tuuri wrote: > > >Mitch, > > > >please send the FULL .err log to me. > > > > > > Hey Heikki, > > Not sure you want that, the file is 44MB uncompressed, and only talks > about the errors reading communication packets. Makes for some really > boring reading ;) The InnoDB error I managed to figure out - I once > upped max_connections without doing the math, and the machine was > exhausting physical RAM. So I backed off max_connections and the really > ugly errors are now gone :) ok, good. It was then a hang caused by excessive OS swapping or something like that. > But I still do not have a solution to my error logs filling up with the > connection error, which is now unfortunately every 1 to 2 seconds... > The machines are connected with gigabit ethernet, and have been running > without trouble for at least three months - and now are filling up the > error logs. Any ideas, anybody? Bueller? I think those ugly warnings will again be removed in the next 4.0 release. They were instroduced to 4.0.xx accidentally. > -- Mitch Regards, Heikki > >Best regards, > > > >Heikki > > > >----- Original Message ----- > >From: "Mitch Pirtle" <[EMAIL PROTECTED]> > >Newsgroups: mailing.database.myodbc > >Sent: Saturday, July 03, 2004 4:41 PM > >Subject: InnoDB and long semaphore waits > > > > > > > > > >>Hi listers, > >> > >>I just got here, so please let me know if this is not the appropriate > >>list! :) > >> > >>Running MySQL 4.0.20 on Fedora Core 1, with InnoDB tables. Installed > >>from RPMs provided at MySQL.com. > >> > >>Last night the beastie came down hard, and requred a physical reboot in > >>order to free/kill some mysqld processes. I say this as I worry that it > >>could be from a hardware error, however it is running on a dual Xeon > >>machine that is not even 6 months old... > >> > >>When I attempted to get the stack trace I only get "nm: > >>/usr/sbin/mysqld: no symbols", and the docs point at stuff that is not > >>on this box. ? > >> > >>I see two errors that need resolution: > >> > >>-------------------------------------------------------------------------- > >> > >> > >------------------- > > > > > >>1) In the error log, I see the following: > >> > >>InnoDB: ###### Diagnostic info printed to the standard error stream > >>InnoDB: Warning: a long semaphore wait: > >>--Thread 23207961 has waited at btr0sea.c line 480 for 625.00 seconds > >>the semaphore: > >>X-lock on RW-latch at 0x4506a768 created in file btr0sea.c line 139 > >>a writer (thread id 23207961) has reserved it in mode wait exclusive > >>number of readers 1, waiters flag 1 > >>Last time read locked in file btr0sea.c line 745 > >>Last time write locked in file btr0sea.c line 480 > >>InnoDB: Error: semaphore wait has lasted > 600 seconds > >>InnoDB: We intentionally crash the server, because it appears to be hung. > >>040702 20:45:27InnoDB: Assertion failure in thread 24583 in file > >>sync0arr.c line 925 > >>InnoDB: We intentionally generate a memory trap. > >>InnoDB: Submit a detailed bug report to http://bugs.mysql.com. > >>InnoDB: If you get repeated assertion failures or crashes, even > >>InnoDB: immediately after the mysqld startup, there may be > >>InnoDB: corruption in the InnoDB tablespace. See section 6.1 of > >>InnoDB: http://www.innodb.com/ibman.php about forcing recovery. > >>mysqld got signal 11; > >> > >>...and goes on to say: > >> > >>key_buffer_size=402653184 > >>read_buffer_size=2093056 > >>max_used_connections=176 > >>max_connections=500 > >>threads_connected=146 > >>It is possible that mysqld could use up to > >>key_buffer_size + (read_buffer_size + sort_buffer_size)*max_connections > >>= 2439212 K > >>bytes of memory > >>Hope that's ok; if not, decrease some variables in the equation. > >> > >>thd=(nil) > >>Attempting backtrace. You can use the following information to find out > >>where mysqld died. If you see no messages after this, something went > >>terribly wrong... > >>Cannot determine thread, fp=0xbfedf758, backtrace may not be correct. > >>Stack range sanity check OK, backtrace follows: > >>0x80720d4 > >>0x8250d48 > >>0x81ed044 > >>0x80f9148 > >>0x824e4fc > >>0x828452a > >>New value of fp=(nil) failed sanity check, terminating stack trace! > >> > >>Again, I try to follow the instructons on dealing with the stack trace > >>but the instructions fail, and also do not provide enough explanation > >>for me to figure out on my own how to fix it :( > >> > >>-------------------------------------------------------------------------- > >> > >> > >------------------- > > > > > >>2) After rebooting the server, the error log fills up with: > >> > >>040703 9:07:59 Aborted connection 55 to db: 'db1' user: 'www' host: > >>`192.168.1.1' (Got an error reading communication packets) > >>...and it repeats itself roughly every 5-to-10 seconds, which I > >>obviously find alarming. > >> > >>Cannot find any reference on that error, and don't really have enough > >>data to know how to react... > >> > >> > >>Can somebody please whack me upside the head with a cluestick? > >> > >>-- Mitch > >> > >>-- > >>MySQL General Mailing List > >>For list archives: http://lists.mysql.com/mysql > >>To unsubscribe: > >> > >> > >http://lists.mysql.com/[EMAIL PROTECTED] > > > > > > > > > > > > > -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
