disk full scenario

if disk is near to full

and i establish a connection with postgres and read a table and then close
the following error logs comes in postgreslog:

17:10:47 CDT%authentication>LOG: connection authorized: user=sdc database=dbexpress
<2006-04-05 17:10:47 CDT%idle>LOG: disconnection: session time: 0:00:00.00 user=sdc database=dbexpress host=[local] port=
<2006-04-05 17:10:47 CDT%idle>LOG: disconnection: session time: 7:41:26.70 user=sdc database=dbexpress host=[local] port=
<2006-04-05 17:10:47 CDT%>ERROR: could not access status of transaction 0
<2006-04-05 17:10:47 CDT%>DETAIL: could not write to file "/export/home1/sdc_image_pool/dbx/pg_clog/0001" at offset 40960:
No space left on device
<2006-04-05 17:10:48 CDT%>LOG: could not close temporary statistics file "/export/home1/sdc_image_pool/dbx/global/pgstat.tm
p.24783": No space left on device
<2006-04-05 17:10:49 CDT%>ERROR: could not access status of transaction 0
<2006-04-05 17:10:49 CDT%>DETAIL: could not write to file "/export/home1/sdc_image_pool/dbx/pg_clog/0001" at offset 40960:
No space left on device
<2006-04-05 17:10:50 CDT%>ERROR: could not access status of transaction 0
<2006-04-05 17:10:50 CDT%>DETAIL: could not write to file "/export/home1/sdc_image_pool/dbx/pg_clog/0001" at offset 40960:
No space left on device
<2006-04-05 17:10:52 CDT%>ERROR: could not access status of transaction 0
<2006-04-05 17:10:52 CDT%>DETAIL: could not write to file "/export/home1/sdc_image_pool/dbx/pg_clog/0001" at offset 40960:
No space left on device

and it keeps logging the same.


and later in the log i also see these messages:
<2006-04-05 23:46:37 CDT%>LOG: could not close temporary statistics file "/export/home1/sdc_image_pool/dbx/global/pgstat.tm
p.8207": No space left on device
<2006-04-05 23:46:47 CDT%>LOG: received fast shutdown request
<2006-04-05 23:46:47 CDT%>LOG: shutting down
<2006-04-05 23:46:47 CDT%>PANIC: could not access status of transaction 0
<2006-04-05 23:46:47 CDT%>DETAIL: could not write to file "/export/home1/sdc_image_pool/dbx/pg_subtrans/0012" at offset 155
648: No space left on device
<2006-04-05 23:46:47 CDT%>LOG: background writer process (PID 8205) was terminated by signal 6
<2006-04-05 23:46:47 CDT%>LOG: terminating any other active server processes
<2006-04-05 23:46:47 CDT%>LOG: all server processes terminated; reinitializing
<2006-04-05 23:46:47 CDT%>LOG: database system shutdown was interrupted at 2006-04-05 23:46:47 CDT
<2006-04-05 23:46:47 CDT%>LOG: checkpoint record is at 0/1FE5D4A0
<2006-04-05 23:46:47 CDT%>LOG: redo record is at 0/1FE5D4A0; undo record is at 0/0; shutdown TRUE
<2006-04-05 23:46:47 CDT%>LOG: next transaction ID: 1217370; next OID: 612902
<2006-04-05 23:46:47 CDT%>LOG: database system was not properly shut down; automatic recovery in progress
<2006-04-05 23:46:47 CDT%>LOG: record with zero length at 0/1FE5D4DC
<2006-04-05 23:46:47 CDT%>LOG: redo is not required
<2006-04-05 23:46:47 CDT%>LOG: database system is ready
<2006-04-05 23:46:47 CDT%>LOG: shutting down
<2006-04-05 23:46:47 CDT%>LOG: database system is shut down
<2006-04-05 23:46:48 CDT%>LOG: could not write "/export/home1/sdc_image_pool/dbx/global/pg_fsm.cache": No space left on dev
ice

<2006-04-05 23:52:03 CDT%>LOG: database system was shut down at 2006-04-05 23:46:47 CDT
<2006-04-05 23:52:03 CDT%>LOG: checkpoint record is at 0/1FE5D518
<2006-04-05 23:52:03 CDT%>LOG: redo record is at 0/1FE5D518; undo record is at 0/0; shutdown TRUE
<2006-04-05 23:52:03 CDT%>LOG: next transaction ID: 1217370; next OID: 612902
<2006-04-05 23:52:03 CDT%>LOG: database system is ready
<2006-04-05 23:52:04 CDT%>LOG: connection received: host=[local] port=
<2006-04-05 23:52:04 CDT%authentication>LOG: connection authorized: user=sdc database=template1
<2006-04-05 23:52:04 CDT%idle>LOG: disconnection: session time: 0:00:00.00 user=sdc database=template1 host=[local] port=
<2006-04-05 23:52:05 CDT%>LOG: connection received: host=[local] port=
<2006-04-05 23:52:05 CDT%authentication>LOG: connection authorized: user=sdc database=dbexpress
<2006-04-05 23:52:05 CDT%idle>LOG: disconnection: session time: 0:00:00.68 user=sdc database=dbexpress host=[local] port=
<2006-04-05 23:52:22 CDT%>LOG: connection received: host=localhost.localdomain port=60013
<2006-04-05 23:52:22 CDT%authentication>LOG: connection authorized: user=sdc database=dbexpress
<2006-04-05 23:53:29 CDT%>LOG: connection received: host=[local] port=
<2006-04-05 23:53:29 CDT%authentication>LOG: connection authorized: user=sdc database=dbexpress
<2006-04-05 23:53:47 CDT%>LOG: connection received: host=localhost.localdomain port=60017
<2006-04-05 23:53:47 CDT%authentication>LOG: connection authorized: user=sdc database=dbexpress
<2006-04-05 23:53:48 CDT%>LOG: could not close temporary statistics file "/export/home1/sdc_image_pool/dbx/global/pgstat.tm
p.10771": No space left on device
<2006-04-05 23:54:02 CDT%>LOG: connection received: host=[local] port=
<2006-04-05 23:54:02 CDT%authentication>LOG: connection authorized: user=sdc database=dbexpress
<2006-04-05 23:54:04 CDT%>LOG: could not close temporary statistics file "/export/home1/sdc_image_pool/dbx/global/pgstat.tm
p.10771": No space left on device



how much is the amt of disk space required in that.

i want to avoid the above process if the left disk space is some x value.

what is the safest value for this x?

thanks,
regards
Surabhi