The file system is ext3. I am calling this api sqlite3_config(SQLITE_CONFIG_LOG, errorLogCallback, NULL);
and I expect the errorLogCallback to be called in error conditions to give more logs/diagnostics on what happened. In this callback as suggested I log the error code and the message. When a sqlite3 api like sqlite3_step returns an error like "file is encrypted or is not a database" , I expect my callback be also called before the sqlite3_Xxx api returns so that I can get more information on what that error happened in the first place. Is my understanding correct since my callback is not getting called ? -Mayank -----Original Message----- From: sqlite-users-boun...@sqlite.org [mailto:sqlite-users-boun...@sqlite.org] On Behalf Of mm.w Sent: Friday, July 11, 2014 7:45 PM To: General Discussion of SQLite Database Subject: Re: [sqlite] Sometimes when my process restarts, it returns error "database is locked" what's your file system? looks like not related at all to sqlite but your code On Fri, Jul 11, 2014 at 4:31 PM, Mayank Kumar (mayankum) <mayan...@cisco.com > wrote: > Hi Richard > I have enabled error logging as suggested. Is there a way to test > this logging. For e.g. I just corrupted the database by modifying the > sqlite db using vi and then got the following message:- > > "file is encrypted or is not a database" > After this error message my process dies(we terminate the process), > and I don't see any logs from my callback. Is this callback called > before the sqlite3_xxx calls returns , if yes then why I don't see any > error messages from the callback ? > > -Mayank > > > -----Original Message----- > From: sqlite-users-boun...@sqlite.org [mailto: > sqlite-users-boun...@sqlite.org] On Behalf Of Richard Hipp > Sent: Tuesday, July 08, 2014 1:05 PM > To: General Discussion of SQLite Database > Subject: Re: [sqlite] Sometimes when my process restarts, it returns > error "database is locked" > > On Tue, Jul 8, 2014 at 3:58 PM, Mayank Kumar (mayankum) < > mayan...@cisco.com> > wrote: > > > Hi All > > > > We have a process with a single sqlite db and a single connection to > > the database from a thread. From time to time our processs restarts > > and reopens the database or tries to create the database if it > > doesn't > exists. > > Sometimes when the process restarts, we get the error "database is > locked" > > although the way the process is restarted is the following:- > > > > -we always starts the process as a child by doing a vfork and exec > > -if we receive a sigchld, we waitpid and finally restarts the > > process in same way > > > > I don't think this is possible. I think something else must be going on. > > Have you enabled logging? See http://www.sqlite.org/errlog.html for > details. The error log might give additional clues. > > > > > > > Could this scenario ever lead to database locked scenario, should we > > build a retry mechanism when we get this error or this shouldn't > > occur if we know the older process has died. > > > > Any suggestions? > > > > -Mayank > > _______________________________________________ > > sqlite-users mailing list > > sqlite-users@sqlite.org > > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > > > > > > -- > D. Richard Hipp > d...@sqlite.org > _______________________________________________ > sqlite-users mailing list > sqlite-users@sqlite.org > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > _______________________________________________ > sqlite-users mailing list > sqlite-users@sqlite.org > http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users > _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
