In <user-5b45d621077f@xymon.invalid> Matthew Moldvan <user-ffaad6c7b6c3@xymon.invalid> writes:
I have a question on monitoring separate filesystems with different thresholds. My hobbit-clients.cfg file on the Xymon server includes the following:
(Names changed to protect the innocent of course).
$MACRO=%^(asdf|qwer|jalopy).*(a|s|h|l).xyz.com
HOST=$MACRO
DISK / 90 95 GROUP=SOMEONES_PAGER
DISK %^/.+$ 90 95 GROUP=SOMEONES_EMAIL
DISK %^/export/data/subdir/subdir2/log.* 95 99
However, Xymon doesn't seem to honor the above DISK entries and still marks both "/export/data/subdir/subdir2/log" and "/export/data/subdir/subdir2/log2" as red when it hits 90%+ due to the DEFAULT entry.
You should reverse the order of your rules. As the hobbit-clients.cfg man page says:
"The entire file is evaluated from the top to bottom, and the first match found is used. So you should put the specific settings first, and the generic ones last.
So what happens is that your second rule "DISK %^/.+$" matches everything,
since all of your filesystems begin with "/", so that is the rule that
is applied.
(Your first rule is OK, because it is not a pattern - so it will only match
the root filesystem).
My questions at this point are:
- Are my regular expressions specified correctly?
Except for the order, yes.
- Does Xymon support full extended regular expressions? (e.g. the difference between grep and egrep on Sun systems)
Xymon uses the "pcre" library, so it support "Perl-compatible" regular
expressions.
- Is my understanding of the above accurate with respect to alerts about file system monitoring being generated on the server?
Not sure what you mean by this, but it seems ok. You must have some corresponding entries in your hobbit-alerts.cfg file for the
"SOMEONES_PAGER" and "SOMEONES_EMAIL" groups, like
GROUP=SOMEONES_EMAIL
MAIL user-d083d77d33f5@xymon.invalid
Regards,
Henrik