DISPLAYGROUP in analysis.cfg
I'm seeing very odd behavior with my analysis.cfg*. My xymon server is running 4.3.28 on Solaris 10.
What I'm seeing is that trying to define a rule with DISPLAYGROUP (or EXDISPLAYGROUP) does not behave as expected from the man-page.
If I have the following in my analysis.cfg:
disk /charlie 23 33 DISPLAYGROUP=Charlie
DISPLAYGROUP=Delta disk /delta 24 34
PAGE=Foo disk /foo 10 20
I get the following from xymond_client --dump-config
DISK /charlie 23% 33% 0 -1 red DISPLAYGROUP=Charlie (line: 373) DISK /delta 24% 34% 0 -1 red (line: 376) DISK /foo 10% 20% 0 -1 red PAGE=Foo (line: 379) UP 3600 -1 (line: 385) LOAD 20.00 60.00 (line: 386) DISK * 90% 95% 0 -1 red (line: 387) INODE * 70% 90% 0 -1 red (line: 388) MEMREAL 100 101 (line: 389) MEMSWAP 50 80 (line: 390) MEMACT 90 97 (line: 391)
Demonstrating that DISPLAYGROUP works when appended to a setting, but does _not_ work when preceding a line with a setting. The DISK setting is accepted, but becomes a global value rather than applying to a subset of the hosts. Yet the PAGE rule does work.
I have run through many scenarios, and see that the following DO work as rules preceding a setting: HOST EXHOST PAGE EXPAGE CLASS EXCLASS TEXT TIME EXTIME and that the following DO NOT work DISPLAYGROUP EXDISPLAYGROUP
I also see that order matters. If I move the PAGE=Foo up in the file:
PAGE=Foo disk /foo 10 20
disk /charlie 23 33 DISPLAYGROUP=Charlie
DISPLAYGROUP=Delta disk /delta 24 34
then the resulting --dump-config reveals some strange parsing:
DISK /foo 10% 20% 0 -1 red PAGE=Foo (line: 374) DISK /charlie 23% 33% 0 -1 red DISPLAYGROUP=Charlie PAGE=Foo (line: 376) DISK /delta 24% 34% 0 -1 red (line: 379) UP 3600 -1 (line: 386) LOAD 20.00 60.00 (line: 387) DISK * 90% 95% 0 -1 red (line: 388) INODE * 70% 90% 0 -1 red (line: 389) MEMREAL 100 101 (line: 390) MEMSWAP 50 80 (line: 391) MEMACT 90 97 (line: 392)
Can anyone help me figure out what's going on? Do I have my analysis.cfg syntax all wrong? Or is there a defect in the parsing?
- It's worth noting that I don't do much with that file. The majority of our clients are old Big Brother clients, so I seldom delve into this file.
-- Do things because you should, not just because you can.
John Thurston 907-465-8591 John.Thurston at alaska.gov Department of Administration State of Alaska
And with more testing, DISPLAYGROUP doesn't seem to work at all.
Group names containing spaces, so enclosed in double-quotes, are shown in the --dump with the leading quote and not the trailing quote.
Group names with only one word appear in the --dump correctly, but are not matched when performing a --test.
The behavior is the same when I use a regex in place of the quoted group name containing spaces.
The only way I can make my analysis.cfg file work is by creating wicked regex definitions of the type HOST
-- Do things because you should, not just because you can.
John Thurston 907-465-8591 John.Thurston at alaska.gov Department of Administration State of Alaska
On 10/16/2019 10:51 AM, John Thurston wrote:
I'm seeing very odd behavior with my analysis.cfg*. My xymon server is running 4.3.28 on Solaris 10.
What I'm seeing is that trying to define a rule with DISPLAYGROUP (or EXDISPLAYGROUP) does not behave as expected from the man-page.
If I have the following in my analysis.cfg:
disk /charlie 23 33 DISPLAYGROUP=Charlie
DISPLAYGROUP=Delta ??????? disk? /delta 24 34
PAGE=Foo ??????? disk /foo 10 20
I get the following from???? xymond_client --dump-config
DISK /charlie 23% 33% 0 -1 red DISPLAYGROUP=Charlie (line: 373) DISK /delta 24% 34% 0 -1 red (line: 376) DISK /foo 10% 20% 0 -1 red PAGE=Foo (line: 379) UP 3600 -1 (line: 385) LOAD 20.00 60.00 (line: 386) DISK * 90% 95% 0 -1 red (line: 387) INODE * 70% 90% 0 -1 red (line: 388) MEMREAL 100 101 (line: 389) MEMSWAP 50 80 (line: 390) MEMACT 90 97 (line: 391)
Demonstrating that DISPLAYGROUP works when appended to a setting, but does _not_ work when preceding a line with a setting.? The DISK setting is accepted, but becomes a global value rather than applying to a subset of the hosts. Yet the PAGE rule does work.
I have run through many scenarios, and see that the following DO work as rules preceding a setting: ? HOST ? EXHOST ? PAGE ? EXPAGE ? CLASS ? EXCLASS ? TEXT ? TIME ? EXTIME and that the following DO NOT work ? DISPLAYGROUP ? EXDISPLAYGROUP
I also see that order matters. If I move the PAGE=Foo up in the file:
PAGE=Foo ??????? disk /foo 10 20
disk /charlie 23 33 DISPLAYGROUP=Charlie
DISPLAYGROUP=Delta ??????? disk? /delta 24 34
then the resulting --dump-config reveals some strange parsing:
DISK /foo 10% 20% 0 -1 red PAGE=Foo (line: 374) DISK /charlie 23% 33% 0 -1 red DISPLAYGROUP=Charlie PAGE=Foo (line: 376) DISK /delta 24% 34% 0 -1 red (line: 379) UP 3600 -1 (line: 386) LOAD 20.00 60.00 (line: 387) DISK * 90% 95% 0 -1 red (line: 388) INODE * 70% 90% 0 -1 red (line: 389) MEMREAL 100 101 (line: 390) MEMSWAP 50 80 (line: 391) MEMACT 90 97 (line: 392)
Can anyone help me figure out what's going on? Do I have my analysis.cfg syntax all wrong? Or is there a defect in the parsing?
- It's worth noting that I don't do much with that file. The majority of our clients are old Big Brother clients, so I seldom delve into this file.
bump
Can anyone say they have seen similar behavior with DISPLAYGROUP in analysis.cfg?
Can anyone say their analysis.cfg works fine with DISPLAYGROUP?
Without DISPLAYGROUP, I'm finding it pretty hard to make my analysis.cfg file manageable.
-- Do things because you should, not just because you can.
John Thurston 907-465-8591 John.Thurston at alaska.gov Department of Administration State of Alaska
On 10/17/2019 6:58 AM, John Thurston wrote:
And with more testing, DISPLAYGROUP doesn't seem to work at all.
Group names containing spaces, so enclosed in double-quotes, are shown in the --dump with the leading quote and not the trailing quote.
Group names with only one word appear in the --dump correctly, but are not matched when performing a --test.
The behavior is the same when I use a regex in place of the quoted group name containing spaces.
The only way I can make my analysis.cfg file work is by creating wicked regex definitions of the type HOST
-- ?? Do things because you should, not just because you can.
John Thurston??? 907-465-8591 John.Thurston at alaska.gov Department of Administration State of Alaska
On 10/16/2019 10:51 AM, John Thurston wrote:
I'm seeing very odd behavior with my analysis.cfg*. My xymon server is running 4.3.28 on Solaris 10.
What I'm seeing is that trying to define a rule with DISPLAYGROUP (or EXDISPLAYGROUP) does not behave as expected from the man-page.
If I have the following in my analysis.cfg:
disk /charlie 23 33 DISPLAYGROUP=Charlie
DISPLAYGROUP=Delta ??????? disk? /delta 24 34
PAGE=Foo ??????? disk /foo 10 20
I get the following from???? xymond_client --dump-config
DISK /charlie 23% 33% 0 -1 red DISPLAYGROUP=Charlie (line: 373) DISK /delta 24% 34% 0 -1 red (line: 376) DISK /foo 10% 20% 0 -1 red PAGE=Foo (line: 379) UP 3600 -1 (line: 385) LOAD 20.00 60.00 (line: 386) DISK * 90% 95% 0 -1 red (line: 387) INODE * 70% 90% 0 -1 red (line: 388) MEMREAL 100 101 (line: 389) MEMSWAP 50 80 (line: 390) MEMACT 90 97 (line: 391)
Demonstrating that DISPLAYGROUP works when appended to a setting, but does _not_ work when preceding a line with a setting.? The DISK setting is accepted, but becomes a global value rather than applying to a subset of the hosts. Yet the PAGE rule does work.
I have run through many scenarios, and see that the following DO work as rules preceding a setting: ?? HOST ?? EXHOST ?? PAGE ?? EXPAGE ?? CLASS ?? EXCLASS ?? TEXT ?? TIME ?? EXTIME and that the following DO NOT work ?? DISPLAYGROUP ?? EXDISPLAYGROUP
I also see that order matters. If I move the PAGE=Foo up in the file:
PAGE=Foo ??????? disk /foo 10 20
disk /charlie 23 33 DISPLAYGROUP=Charlie
DISPLAYGROUP=Delta ??????? disk? /delta 24 34
then the resulting --dump-config reveals some strange parsing:
DISK /foo 10% 20% 0 -1 red PAGE=Foo (line: 374) DISK /charlie 23% 33% 0 -1 red DISPLAYGROUP=Charlie PAGE=Foo (line: 376) DISK /delta 24% 34% 0 -1 red (line: 379) UP 3600 -1 (line: 386) LOAD 20.00 60.00 (line: 387) DISK * 90% 95% 0 -1 red (line: 388) INODE * 70% 90% 0 -1 red (line: 389) MEMREAL 100 101 (line: 390) MEMSWAP 50 80 (line: 391) MEMACT 90 97 (line: 392)
Can anyone help me figure out what's going on? Do I have my analysis.cfg syntax all wrong? Or is there a defect in the parsing?
- It's worth noting that I don't do much with that file. The majority of our clients are old Big Brother clients, so I seldom delve into this file.
participants (1)
-
john.thurston＠alaska.gov