Alert Config.

5 messages in this thread

list James Wade · Tue, 30 Oct 2007 14:46:37 -0500 ·

Will this work O.K.?

 
HOST=%ep* SERVICE=http

MAIL user-5987f8432f17@xymon.invalid REPEAT=15 COLOR=RED DURATION>5m DURATION<35
RECOVERED

 
So, that it will not page the cell unless the outage lasts longer than 5
minutes,

and will only page twice. (Assuming after 5 minute outage it pages, then
pages

every fifteen minutes for thirty minutes - twice)

 
Thanks..James

 
James Wade

list Henrik Størner · Tue, 30 Oct 2007 22:07:41 +0100 ·

On Tue, Oct 30, 2007 at 02:46:37PM -0500, James Wade wrote:

Will this work O.K.?

 
HOST=%ep* SERVICE=http

the regex is wrong, you probably want "%ep.*"

▸ quoted from James Wade

MAIL user-5987f8432f17@xymon.invalid REPEAT=15 COLOR=RED DURATION>5m DURATION<35 RECOVERED

So, that it will not page the cell unless the outage lasts longer than 5
minutes,

and will only page twice. (Assuming after 5 minute outage it pages, then
pages every fifteen minutes for thirty minutes - twice)

For http, this will work as intended. There is a "gotcha" when this is
used with tests that can go yellow - this is not the case with "http"
tests. But e.g. a disk check that goes yellow for an hour, then red
would NOT trigger an alert in this case, first because you don't alert
on yellow, and second because the duration of the incident exceeds 35
minutes when it does go red (DURATION begins when the status goes
non-green).


Regards,
Henrik

list Josh Luthman · Tue, 30 Oct 2007 18:28:15 -0400 ·

While we're on the subject of alert configuration, could someone confirm
mine?

HOST=*
  MAIL user-4c45a83f15cb@xymon.invalid COLOR=RED DURATION>15m RECOVERED

It will only email once something has been red for 15m and it will email
once it has become green again - correct?

▸ quoted from Henrik Størner


On 10/30/07, Henrik Stoerner <user-ce4a2c883f75@xymon.invalid> wrote:

On Tue, Oct 30, 2007 at 02:46:37PM -0500, James Wade wrote:

Will this work O.K.?


HOST=%ep* SERVICE=http

the regex is wrong, you probably want "%ep.*"

MAIL user-5987f8432f17@xymon.invalid REPEAT=15 COLOR=RED DURATION>5m DURATION<35

RECOVERED

So, that it will not page the cell unless the outage lasts longer than 5
minutes,

and will only page twice. (Assuming after 5 minute outage it pages, then
pages every fifteen minutes for thirty minutes - twice)

For http, this will work as intended. There is a "gotcha" when this is
used with tests that can go yellow - this is not the case with "http"
tests. But e.g. a disk check that goes yellow for an hour, then red
would NOT trigger an alert in this case, first because you don't alert
on yellow, and second because the duration of the incident exceeds 35
minutes when it does go red (DURATION begins when the status goes
non-green).


Regards,
Henrik

--


Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX

Those who don't understand UNIX are condemned to reinvent it, poorly.
--- Henry Spencer

list Allan Spencer · Wed, 31 Oct 2007 10:05:42 +1100 ·

▸ quoted from Josh Luthman

Josh Luthman wrote:

While we're on the subject of alert configuration, could someone 
confirm mine?

HOST=*
  MAIL user-4c45a83f15cb@xymon.invalid


<mailto:user-4c45a83f15cb@xymon.invalid> COLOR=RED DURATION>15m RECOVERED

▸ quoted from Josh Luthman


It will only email once something has been red for 15m and it will 
email once it has become green again - correct?

On 10/30/07, *Henrik Stoerner* < user-ce4a2c883f75@xymon.invalid 
<mailto:user-ce4a2c883f75@xymon.invalid>> wrote:

    On Tue, Oct 30, 2007 at 02:46:37PM -0500, James Wade wrote:

Will this work O.K.?


HOST=%ep* SERVICE=http

    the regex is wrong, you probably want "%ep.*"

MAIL user-5987f8432f17@xymon.invalid <mailto:user-5987f8432f17@xymon.invalid>

▸ quoted from Josh Luthman

    REPEAT=15 COLOR=RED DURATION>5m DURATION<35 RECOVERED

So, that it will not page the cell unless the outage lasts
    longer than 5
minutes,

and will only page twice. (Assuming after 5 minute outage it
    pages, then
pages every fifteen minutes for thirty minutes - twice)

    For http, this will work as intended. There is a "gotcha" when this is
    used with tests that can go yellow - this is not the case with "http"
    tests. But e.g. a disk check that goes yellow for an hour, then red
    would NOT trigger an alert in this case, first because you don't alert
    on yellow, and second because the duration of the incident exceeds 35
    minutes when it does go red (DURATION begins when the status goes
    non-green).


    Regards,
    Henrik


-- 
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX

Those who don't understand UNIX are condemned to reinvent it, poorly.
--- Henry Spencer


Correct :)

list Josh Luthman · Tue, 30 Oct 2007 20:33:24 -0400 ·

Greatly appreciate it.  Thanks Allan!

▸ quoted from Allan Spencer


On 10/30/07, Allan <user-42a3456c44ef@xymon.invalid> wrote:

Josh Luthman wrote:

While we're on the subject of alert configuration, could someone
confirm mine?

HOST=*
  MAIL user-4c45a83f15cb@xymon.invalid
<mailto:user-4c45a83f15cb@xymon.invalid> COLOR=RED DURATION>15m RECOVERED

It will only email once something has been red for 15m and it will
email once it has become green again - correct?

On 10/30/07, *Henrik Stoerner* < user-ce4a2c883f75@xymon.invalid
<mailto:user-ce4a2c883f75@xymon.invalid>> wrote:

    On Tue, Oct 30, 2007 at 02:46:37PM -0500, James Wade wrote:

Will this work O.K.?


HOST=%ep* SERVICE=http

    the regex is wrong, you probably want "%ep.*"

MAIL user-5987f8432f17@xymon.invalid <mailto:user-5987f8432f17@xymon.invalid>

    REPEAT=15 COLOR=RED DURATION>5m DURATION<35 RECOVERED

So, that it will not page the cell unless the outage lasts
    longer than 5
minutes,

and will only page twice. (Assuming after 5 minute outage it
    pages, then
pages every fifteen minutes for thirty minutes - twice)

    For http, this will work as intended. There is a "gotcha" when this
is
    used with tests that can go yellow - this is not the case with

"http"

    tests. But e.g. a disk check that goes yellow for an hour, then red
    would NOT trigger an alert in this case, first because you don't
alert
    on yellow, and second because the duration of the incident exceeds

    minutes when it does go red (DURATION begins when the status goes
    non-green).


    Regards,
    Henrik


--
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX

Those who don't understand UNIX are condemned to reinvent it, poorly.
--- Henry Spencer


Correct :)

-- 
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX

Those who don't understand UNIX are condemned to reinvent it, poorly.
--- Henry Spencer

Alert Config. 🔗 link

Alert Config.