Hi Leonardo, Thanks for your suggestion. I changed the two entries in my.cnf to point to the correct folder (mysql-5.5.20-linux2.6-x86_64/data) along with deletion of two log files from data folder and mysql was able to launch.
Jayneel > Hi Leonardo, > > I tried to use the my.cnf from > apache-olio-php-src-0.2/webapp/php/trunk/etc/. But the mysql will not > launch using the new my.cnf file. The error log says: > > 120708 03:33:49 mysqld_safe Starting mysqld daemon with databases from > /root/jayneel/cloudsuite/webserving/web-release/mysql-5.5.20-linux2.6-x86_64/data > 120708 3:33:49 [Note] Plugin 'FEDERATED' is disabled. > 120708 3:33:49 [Warning] option 'innodb-autoextend-increment': unsigned > value 104857600 adjusted to 1000 > 120708 3:33:49 InnoDB: The InnoDB memory heap is disabled > 120708 3:33:49 InnoDB: Mutexes and rw_locks use GCC atomic builtins > 120708 3:33:49 InnoDB: Compressed tables use zlib 1.2.3 > 120708 3:33:49 InnoDB: Using Linux native AIO > 120708 3:33:49 InnoDB: Initializing buffer pool, size = 2.0G > 120708 3:33:49 InnoDB: Completed initialization of buffer pool > 120708 3:33:49 InnoDB: highest supported file format is Barracuda. > InnoDB: No valid checkpoint found. > InnoDB: If this error appears when you are creating an InnoDB database, > InnoDB: the problem may be that during an earlier attempt you managed > InnoDB: to create the InnoDB data files, but log file creation failed. > InnoDB: If that is the case, please refer to > InnoDB: http://dev.mysql.com/doc/refman/5.5/en/error-creating-innodb.html > 120708 3:33:49 [ERROR] Plugin 'InnoDB' init function returned error. > 120708 3:33:49 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE > failed. > 120708 3:33:49 [ERROR] Unknown/unsupported storage engine: innodb > 120708 3:33:49 [ERROR] Aborting > > 120708 3:33:49 [Note] > /root/jayneel/cloudsuite/webserving/web-release/mysql-5.5.20-linux2.6-x86_64/bin/mysqld: > Shutdown complete > > 120708 03:33:49 mysqld_safe mysqld from pid file > /root/jayneel/cloudsuite/webserving/web-release/mysql-5.5.20-linux2.6-x86_64/data/sc-h03.cs.wisc.edu.pid > ended > > I am not sure what is the trouble with InnoDB settings. Did you make any > changes to the my.cnf file? > > Thanks, > Jayneel > >> I think so, I used the same file and after 72 hours I had not been able >> to >> load the database. I used the one pointed to Cansu and it took just some >> hours. >> >> On Saturday, July 7, 2012, Jayneel Gandhi wrote: >> >>> Hi Leonardo, >>> >>> The mysql conf file was copied from >>> >>> cp mysql-5.5.20-linux2.6-x86_64/**support-files/my-medium.cnf >>> /etc/my.cnf >>> >>> as per the instructions on the webpage. In that conf file innodb plugin >>> is >>> not enabled at all. Am I using the wrong conf file? >>> >>> >>> Thanks, >>> Jayneel >>> >>> On 07/07/2012 10:26 PM, Leonardo Piga wrote: >>> >>>> Cansu answered this question some time ago. It did solve my problem. >>>> Maybe you are having the same issue. >>>> >>>> >>>> Dear Leonardo, >>>> Most probably, this is because of the disk write latency that the >>>> insertion operations are exposed to. You can verify it by checking the >>>> vmstat output, the 'wa' field. >>>> If the wait time is high, please compare your MySQL configuration >>>> parameters with the ones in my.cnf, under >>>> apache-olio-php-src-0.2/**webapp/php/trunk/etc/. >>>> In particular, make sure that the buffer pool and log buffer are large >>>> (need tuning depending on the aggregate memory size). >>>> Moreover, make sure that innodb_doublewrite and >>>> innodb_flush_log_at_trx_commit parameters are set to 0 and 2 >>>> respectively as in my.cnf, so that the database is exposed to the disk >>>> latency at the minimum. >>>> >>>> >>>> >>>> Leonardo >>>> >>>> >>>> On Sat, Jul 7, 2012 at 10:22 PM, Jayneel Gandhi<[email protected]> >>>> wrote: >>>> >>>>> Hi, >>>>> >>>>> The script dbloader.sh takes a lot of time (script has been running >>>>> for >>>>> hours) to load to database even with a load factor of 100. Can anyone >>>>> comment on what should be used as the load factor in case we do want >>>>> to >>>>> simulate for larger number of concurrent users (maybe ~100s or >>>>> ~1000s)? >>>>> Also, what kind of runtime should I expect to load the database in >>>>> that >>>>> case. >>>>> >>>>> Thanks, >>>>> Jayneel >>>>> >>>> >>> >> >> -- >> Leonardo >> > >
