2

Usually when there is a problem, the filesystem mounts itself read-only:

[449551.919049] EXT3-fs (dm-2): error: ext3_journal_start_sb: Detected aborted journal
[449551.919056] EXT3-fs (dm-2): error: remounting filesystem read-only

Now I tried to catch that using monitoring via SNMP that we have in place anyway, but it didn't:

# snmpwalk -c public -v 2c localhost HOST-RESOURCES-MIB::hrFSTable
SNMP table: HOST-RESOURCES-MIB::hrFSTable

...
HOST-RESOURCES-MIB::hrFSAccess.11 = INTEGER: readWrite(1)
HOST-RESOURCES-MIB::hrFSAccess.12 = INTEGER: readWrite(1) ***

It seems to show the configured, not the current state.

Is there any other way to catch this, preferably using SNMP, without resorting to actually having to write a file and check the outcome?

BTW

  • OS: SLES11-SP3
  • Kernel 3.0.101
  • net-snmp-5.4.2.1
Marki
  • 2,795
  • 3
  • 27
  • 45
  • Combination of extension command in snmpd.conf and recipies from here (https://serverfault.com/questions/193971/determine-if-filesystem-or-partition-is-mounted-ro-or-rw-via-bash-script) should do the trick. – Yuri Lachin Nov 16 '18 at 14:44
  • Wouldn't you like to add this as an answer? My comment and follow-up to that answer will be why snmp doesn't do that out-of-the-box :) – Marki Nov 16 '18 at 18:22

1 Answers1

1

Combination of extension command in snmpd.conf and recipies from here (serverfault) should do the trick.

Yuri Lachin
  • 176
  • 4