Database corupption on pg 7.3.3

Hi,
I have had two times problem with pg database corruption.
My system:

Dual athlon 2.4
1 GB RAM,
linux red hat 7.3
reiserFS

Also on the system I have cron that every 3 h. make dump and every 24 h.
make vacuum full analyze.

After working some days pg crashes and I was not able to restore it.

I found tis messages into the log:
Lock file "/mnt/diske/skladdb/postmaster.pid" already exists.
Is another postmaster (pid 1208) running in "/mnt/diske/skladdb"?
LOG:  database system was interrupted at 2003-08-22 20:10:51 EEST
LOG:  checkpoint record is at 0/1829A1E4
LOG:  redo record is at 0/1829A1E4; undo record is at 0/0; shutdown
FALSE
LOG:  next transaction id: 45726; next oid: 13875914
LOG:  database system was not properly shut down; automatic recovery in
progress
LOG:  redo starts at 0/1829A224
PANIC:  Invalid page header in block 4 of 16645
LOG:  startup process (pid 1212) was terminated by signal 6
LOG:  aborting startup due to startup process failure
LOG:  database system was interrupted being in recovery at 2003-08-25
09:03:14 EEST
        This probably means that some data blocks are corrupted
        and you will have to use the last backup for recovery.
LOG:  checkpoint record is at 0/1829A1E4
LOG:  redo record is at 0/1829A1E4; undo record is at 0/0; shutdown
FALSE
LOG:  next transaction id: 45726; next oid: 13875914
LOG:  database system was not properly shut down; automatic recovery in
progress
LOG:  redo starts at 0/1829A224
PANIC:  Invalid page header in block 4 of 16645
LOG:  startup process (pid 1211) was terminated by signal 6
LOG:  aborting startup due to startup process failure
LOG:  database system was interrupted being in recovery at 2003-08-25
09:10:29 EEST
        This probably means that some data blocks are corrupted
        and you will have to use the last backup for recovery.
LOG:  checkpoint record is at 0/1829A1E4
LOG:  redo record is at 0/1829A1E4; undo record is at 0/0; shutdown
FALSE
LOG:  next transaction id: 45726; next oid: 13875914
LOG:  database system was not properly shut down; automatic recovery in
progress
LOG:  redo starts at 0/1829A224
PANIC:  Invalid page header in block 4 of 16645
LOG:  startup process (pid 1211) was terminated by signal 6
LOG:  aborting startup due to startup process failure

Pls, help me to solve this problem.

regards,
ivan.