Hi Stephan, Thanks for the extra pointers. It appears the issue is with the amount of SQL cache entries - the default value is not enough for your deployment. This is generating 'unscheduled' purges to the MySQL database, where the software doesn't try to create the table (will look into this). From the debug you can reckon these ocurrences from the '(CAUSE: safe action)'.
Solution is to increase the default sql_cache_entries value (32771). It is adviceable you pick a prime number. I suggest to increase by one order of magnitude at a time and stop until you see those messages disappearing from the debug. Let me know how it goes. Cheers, Paolo On Thu, Apr 28, 2011 at 07:34:02PM +0200, Stephan Mol | Atrato IP Networks wrote: > Hi Paolo, > > Thank you for your reply. I'm using version 0.14.0rc1. I linked the two files > in the previous mail, something must have gone wrong. The URL's are: > > Configuration: http://noc.as5580.net/~stephan/sfacctd.html > > Sql table schema: http://noc.as5580.net/~stephan/acct_bgp_5mins.html > > Thanks in advance. > > Kind regards, > > Stephan > > On 28 apr 2011, at 17:19, Paolo Lucente wrote: > > > Hi Stephan, > > > > I'd like to know what version you are using and whether you can post your > > configuration (can't spot it in your original email) and the content of the > > file pointed by the sql_table_schema directive. > > > > Cheers, > > Paolo > > > > On Thu, Apr 28, 2011 at 10:41:47AM +0200, Stephan Mol | Atrato IP Networks > > wrote: > >> Dear Paolo, all, > >> > >> I hope you can help me with a problem I'm facing with pmacct storing data > >> in MySQL. Pmacct is setup with a 5 minute cycle, on every 5th minute it > >> tries to store the collected data into MySQL but it runs into this error > >> every time: > >> > >> ( 5mins/mysql ) *** Purging cache - START *** > >> ERROR ( 5mins/mysql ): PRIMARY 'mysql' backend trouble. > >> ERROR ( 5mins/mysql ): The SQL server says: Table > >> 'pmacct.acct_bgp_5mins_20110428_10-15' doesn't exist > >> ( 5mins/mysql ) *** Purging cache - END (QN: 0, ET: 0) *** > >> > >> Then on the 6th minute (where the next collection cycle begins) it tries > >> to store the data again, and this time the table is created and data is > >> stored into it. > >> > >> Sometimes pmacct runs into this error several times, eventually it > >> succeeds storing the data. > >> > >> I'm worried not all the data is being stored because of these errors. How > >> can I resolve this behavior, what is going on? > >> > >> Here's the pmacct config and table file I use. > >> > >> If there's any more info I need to provide please let me know. > >> > >> Thanks in advance. > >> > >> Kind regards, > >> > >> Stephan Mol > >> > >> [ ... ] > > > > > > > > _______________________________________________ > > pmacct-discussion mailing list > > http://www.pmacct.net/#mailinglists > _______________________________________________ pmacct-discussion mailing list http://www.pmacct.net/#mailinglists
