I know, I've been trying everything the past few hours. I've dropped and remade the tables a whole bunch. Spontaneously, there is progress. I'm now getting a different error:
Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: gettablecount(0): no such table: t_lfn (1) Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: gettablecount(1): no such table: t_pfn (1) Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: gettablecount(2): no such table: t_map (1) Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: update_readrli: DB error: no such table: t_rli (1) Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: gettablecount(3): no such table: t_lfn (1) Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: gettablecount(4): no such table: t_lrc (1) Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: gettablecount(5): no such table: t_sender (1) Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: gettablecount(6): no such table: t_map (1) Apr 24 12:09:12 asparagine globus-rls-server[16751]: T3085833920: update_readrli: DB error: no such table: t_rli (1) Also, I was surprised to see that a different rls-server on a different machine that is tied into the same postgres installation (differently named DBs) has remained functional all this time, and seems to function even without postmaster running. There is data in it, but I'm not convinced that data is even in the database: [EMAIL PROTECTED]:/export/work/globus-4.0.6> bin/globus-rls-admin -S rlsn://valine Version: 4.6 Uptime: 00:18:21 LRC stats update method: lfnlist update method: bloomfilter updates lfnlist: rlsn://valine.umiacs.umd.edu:39281 last 04/24/08 11:47:11 lfnlist update interval: 86400 bloomfilter update interval: 900 numlfn: 15 numpfn: 27 nummap: 27 RLI stats updated by: rlsn://valine.umiacs.umd.edu:39281 last 04/24/08 11:47:11 updated via lfnlists numlfn: 15 numlrc: 1 numsender: 1 nummap: 15 Does that have something to with this bloomfilter stuff? I thought since I configured the RLI and the LRC together that the bloomfilter method wouldn't be used. Any little tips are appreciated! Thanks, Adam On Thu, Apr 24, 2008 at 12:06 PM, Robert Schuler <[EMAIL PROTECTED]> wrote: > I haven't seen this before. I know its possible with _SQLite_ due to > exclusive write locking, but I am surprised you would run into this with > _Postgres_. I can't even find that error message for postgres. > > > -----Original Message----- > From: [EMAIL PROTECTED] on behalf of Adam Bazinet > Sent: Wed 4/23/2008 8:48 PM > To: GT User > Subject: [gt-user] RLS woes: globus_io_register_read() timed out > > To anyone who works on RLS, I keep getting this error all of a sudden: > globus_rls_client: IO timeout: globus_io_register_read() timed out after 30 > seconds > On any query of any kind. I even dropped and recreated the underlying > database tables. Seeing a few different things in /var/log/messages: > > Apr 23 23:31:39 asparagine globus-rls-server[11232]: T3086554816: > gettablecount(4): database is locked (5) > Apr 23 23:33:19 asparagine globus-rls-server[11232]: T3086554816: > gettablecount(5): database is locked (5) > Apr 23 23:34:59 asparagine globus-rls-server[11232]: T3086554816: > gettablecount(6): database is locked (5) > Apr 23 23:37:19 asparagine globus-rls-server[11232]: T2929064880: > write_update(rlsn://asparagine.umiacs.umd.edu:39281): globus_rls_client: > IO > timeout: globus_io_register_read() timed out after 120 seconds > Apr 23 23:39:59 asparagine globus-rls-server[11232]: T3086554816: > update_readrli: DB error: database is locked (5) > Apr 23 23:40:00 asparagine globus-rls-server[11232]: T3076062128: readcb: > an > end-of-file was reached globus_xio: An end of file occurred > > Tried googling the database is locked thing, keep restarting postgres, > rebooting the machine... anything else I can try? Not sure how it could > stop working all of a sudden. > > Thanks, > Adam > >
