Re: [GENERAL] Data corruption zero a file - help!!

[Noel Faux]

Thanks for the pointers Michael! Which config file will tell us how big the bock sizes are? Cheers Noel Michael Fuhr wrote: On Wed, Mar 01, 2006 at 04:12:53PM +1100, Noel Faux wrote: Now after doing some searches I managed to work out that the data corruption starts at 902292.137 using this sql: SELECT * FROM gap WHERE ctid = '(902292,$x)' Where $x I changed from 1-150. as mentioned on this post:http://archives.postgresql.org/pgsql-general/2005-11/msg01148.php Following this post it seems all we need to do is re-zero from this point on. However we're not sure which file to do this in. This earlier message in that thread should be helpful: http://archives.postgresql.org/pgsql-general/2005-11/msg01141.php I've worked out the database/relation files are $PGDATA/37958/111685332.* with the max * being 101. Is your table really over 100G? Anyway, if the block size is 8192 then 902292 sould be in the .6 file. If you can spare the time then you might run the dd and od commands that Tom Lane mentions in the above message and post the output. I think the command would be dd bs=8k skip=115860 count=1 if=$PGDATA/37958/111685332.6 | od -x See Tom's message for how I arrived at .6 and 115860 (and verify the math yourself).

begin:vcard
fn:Noel Faux
n:Faux;Noel
org:Monash University;Biochemistry and Molecular Biology
adr:;;;Clayton;Vic;3800;Australia
email;internet:[EMAIL PROTECTED]
tel;work:+61 03 9905 1418
url:http://vbc.med.monash.edu.au/~fauxn
version:2.1
end:vcard


---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
       choose an index scan if your joining column's datatypes do not
       match