[BUGS] BUG #8238: duplicate of bug #6372 on panffs
Sorry to replay in this way but I just joined the list based on a thread I found from a few days ago. I hit the panfs fsync problem this morning and it appears the same as what Jim Hughes is reporting. Initially thought it was SELinux (RHEL 6.4) but it persisted after I turned it off. Bug #6372 and Jim's reply reference a patch for copydir.c but I don't see a patch. It might have been an attachment that I didn't see through the archives. Is this patch available with instructions for compilation? Thanks, Kris -- Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-bugs
Re: [BUGS] BUG #8238: duplicate of bug #6372 on panffs
james.e.hug...@boeing.com writes: Upon creating a database on a panfs (Panasas proprietary fs) directory I get: createdb: database creation failed: ERROR: could not fsync file base/16387: Invalid argument This is a duplicate of bug #6372 and is fixed by applying the patch for copydir.c listed in the thread. So we confirm this problem also exists on another fs in addition to CIFS. Thanks for the report. Looking back at the previous thread, it seems we chose not to change the code (to ignore EINVAL for fsync-on-a-directory) on the grounds that CIFS couldn't be considered a supported filesystem anyway. Do you find that panfs works all right for Postgres other than this issue? regards, tom lane -- Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-bugs
[BUGS] BUG #8238: duplicate of bug #6372 on panffs
The following bug has been logged on the website: Bug reference: 8238 Logged by: Jim Hughes Email address: james.e.hug...@boeing.com PostgreSQL version: 9.1.4 Operating system: SLES 11 Description: Upon creating a database on a panfs (Panasas proprietary fs) directory I get: createdb: database creation failed: ERROR: could not fsync file base/16387: Invalid argument This is a duplicate of bug #6372 and is fixed by applying the patch for copydir.c listed in the thread. So we confirm this problem also exists on another fs in addition to CIFS. -- Sent via pgsql-bugs mailing list (pgsql-bugs@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-bugs