Jan-Henrik Haukeland wrote:
On 23. jan. 2006, at 21.35, Martin Pala wrote:
Maybe we can add generic 'retry' argument to the action to set the
count of actions performed when the service remains in the same
state. Something like:
if failed then <action1> [retry <x> [times]]
[else if passed then <action2> [retry <y> [times]]]
Nice suggestion! +1
- add the 'retry' option to the <action> (value of -1 means no limits)
Or if not [retry] is used, defaults to no limits.
- try the action only once by default (even in the case of failed
state)
Retry is better, for a process you may want monit to try and (re)
start it on each poll-cycle until timeout. I just saw that with retry
we do not need the timeout statement. Hmm :) Retry is a new paradigm
which means we can refactor away timeout(?)
would'nt there be a difference in the way retry and timeout behave?
- if you use retry, the service would still be monitored while down
(after x retry times)
(this is what i need :-)
- if you use timeout, the service would not be monitored after timeout
so retry and timeout are usefull
christian
--
Jan-Henrik Haukeland
Mobil +47 97141255
--
To unsubscribe:
http://lists.nongnu.org/mailman/listinfo/monit-general
begin:vcard
fn:Christian Kowarzik
n:Kowarzik;Christian
org:PROFESSIONAL IT-SERVICE
adr;quoted-printable;quoted-printable:;;Kaiserstra=C3=9Fe 28b;D=C3=BCsseldorf;NRW;40479;Deutschland
email;internet:[EMAIL PROTECTED]
tel;work:0211 - 545 11 37
tel;fax:0211 - 152 00 50
tel;cell:0179 - 226 28 90
x-mozilla-html:FALSE
version:2.1
end:vcard
--
To unsubscribe:
http://lists.nongnu.org/mailman/listinfo/monit-general
