2 graphing questions
list Adam Scheblein
Greetings, I know hobbit is sent vmstat info which has page in/out information - is it possible to make a vmstat2 graph which has page outs and a vmstat3 graph which has page ins (or something like that) and if most of the code is already in place, what do I have to change to make this work?? My next question is when I made a change to my hobbitserver.cfg so that in TEST2RRD cpu=vmstat when I look in trends at load (la) and clock, they have stopped graphing. Why is this happening, and is there anything I can do to fix it?? Thanks, Adam Scheblein
list Jerry Yu
1) they are already defined in '/etc/hobbit/hobbitgraph.cfg', including PI/PO/CS, etc. The one included under 'trends' is 'vmstat'. 2) I believe you need to add this 'vmstat3' to GRAPHS= line inside hobbitserver.cfg revert your change to have 'cpu=la' & you may have to remove the corrupted rrd databases as well.
▸
On 9/7/06, Scheblein, Adam <user-de8d51f0c651@xymon.invalid> wrote:Greetings, I know hobbit is sent vmstat info which has page in/out information – is it possible to make a vmstat2 graph which has page outs and a vmstat3 graph which has page ins (or something like that) and if most of the code is already in place, what do I have to change to make this work?? My next question is when I made a change to my hobbitserver.cfg so that in TEST2RRD cpu=vmstat when I look in trends at load (la) and clock, they have stopped graphing… Why is this happening, and is there anything I can do to fix it?? Thanks, Adam Scheblein
list Johann Eggers
Hi, the graph definition is all in the hobbitgraph.cfg You can change/extend it to fit to your needs... Johann
▸
From: Jerry Yu [mailto:user-764c1f364fe0@xymon.invalid]
Sent: Donnerstag, 7. September 2006 16:44
To: user-ae9b8668bcde@xymon.invalid
Subject: ***** POSSIBLE SPAM ***** Re: [hobbit] 2 graphing questions
1) they are already defined in '/etc/hobbit/hobbitgraph.cfg', including PI/PO/CS, etc. The one included under 'trends' is 'vmstat'.
2) I believe you need to add this 'vmstat3' to GRAPHS= line inside hobbitserver.cfg
revert your change to have 'cpu=la' & you may have to remove the corrupted rrd databases as well.
On 9/7/06, Scheblein, Adam <user-de8d51f0c651@xymon.invalid> wrote:
Greetings,
I know hobbit is sent vmstat info which has page in/out information - is it possible to make a vmstat2 graph which has page outs and a vmstat3 graph which has page ins (or something like that) and if most of the code is already in place, what do I have to change to make this work??
My next question is when I made a change to my hobbitserver.cfg so that in TEST2RRD cpu=vmstat when I look in trends at load (la) and clock, they have stopped graphing... Why is this happening, and is there anything I can do to fix it??
Thanks,
Adam Scheblein
list Adam Scheblein
1) [solved] OK -- I should have looked harder... I needed vmstat8 2) When I put things in the "graphs" section it only changes the trends page. The whole reason that I originally made the change was so that when I click into the cpu info page, I would get the vmstat graph, however, I still wanted la and clock to continue graphing on the trends page. When I change CPU back to la both la and clock start graphing again, so I know the db's/rrd's are not corrupt. So 1 solved, 1 to go :) Thanks, Adam
▸
-----Original Message----- From: Johann Eggers [mailto:user-769b09132207@xymon.invalid] Sent: Thursday, September 07, 2006 10:43 AM To: user-ae9b8668bcde@xymon.invalid Subject: [hobbit] RE: ***** POSSIBLE SPAM ***** Re: [hobbit] 2 graphing questions Hi, the graph definition is all in the hobbitgraph.cfg You can change/extend it to fit to your needs... Johann From: Jerry Yu [mailto:user-764c1f364fe0@xymon.invalid] Sent: Donnerstag, 7. September 2006 16:44 To: user-ae9b8668bcde@xymon.invalid Subject: ***** POSSIBLE SPAM ***** Re: [hobbit] 2 graphing questions 1) they are already defined in '/etc/hobbit/hobbitgraph.cfg', including PI/PO/CS, etc. The one included under 'trends' is 'vmstat'. 2) I believe you need to add this 'vmstat3' to GRAPHS= line inside hobbitserver.cfg revert your change to have 'cpu=la' & you may have to remove the corrupted rrd databases as well. On 9/7/06, Scheblein, Adam <user-de8d51f0c651@xymon.invalid> wrote: Greetings, I know hobbit is sent vmstat info which has page in/out information - is it possible to make a vmstat2 graph which has page outs and a vmstat3 graph which has page ins (or something like that) and if most of the code is already in place, what do I have to change to make this work?? My next question is when I made a change to my hobbitserver.cfg so that in TEST2RRD cpu=vmstat when I look in trends at load (la) and clock, they have stopped graphing... Why is this happening, and is there anything I can do to fix it?? Thanks, Adam Scheblein
list Adam Scheblein
1) [solved] OK -- I should have looked harder... I needed vmstat8 2) When I put things in the "graphs" section it only changes the trends page. The whole reason that I originally made the change was so that when I click into the cpu info page, I would get the vmstat graph, however, I still wanted la and clock to continue graphing on the trends page. When I change CPU back to la both la and clock start graphing again, so I know the db's/rrd's are not corrupt. So 1 solved, 1 to go :) Thanks, Adam
-----Original Message----- From: Johann Eggers [mailto:user-769b09132207@xymon.invalid] Sent: Thursday, September 07, 2006 10:43 AM To: user-ae9b8668bcde@xymon.invalid Subject: [hobbit] RE: ***** POSSIBLE SPAM ***** Re: [hobbit] 2 graphing questions Hi, the graph definition is all in the hobbitgraph.cfg You can change/extend it to fit to your needs... Johann From: Jerry Yu [mailto:user-764c1f364fe0@xymon.invalid] Sent: Donnerstag, 7. September 2006 16:44 To: user-ae9b8668bcde@xymon.invalid Subject: ***** POSSIBLE SPAM ***** Re: [hobbit] 2 graphing questions 1) they are already defined in '/etc/hobbit/hobbitgraph.cfg', including PI/PO/CS, etc. The one included under 'trends' is 'vmstat'. 2) I believe you need to add this 'vmstat3' to GRAPHS= line inside hobbitserver.cfg revert your change to have 'cpu=la' & you may have to remove the corrupted rrd databases as well. On 9/7/06, Scheblein, Adam <user-de8d51f0c651@xymon.invalid> wrote: Greetings, I know hobbit is sent vmstat info which has page in/out information - is it possible to make a vmstat2 graph which has page outs and a vmstat3 graph which has page ins (or something like that) and if most of the code is already in place, what do I have to change to make this work?? My next question is when I made a change to my hobbitserver.cfg so that in TEST2RRD cpu=vmstat when I look in trends at load (la) and clock, they have stopped graphing... Why is this happening, and is there anything I can do to fix it?? Thanks, Adam Scheblein
list T.J. Yang
I got "service not avaiable" in my web browser all the time. but if I refresh my web browser, the page will showup again. looking at bb2.html, found bbgen turn yellow to green reqularly. "hobbitd status-board not available" showen in bbgen's output. I did search hobbit mail archive and found following related thread, http://www.hswn.dk/hobbiton/2005/06/msg00186.html I added --debug when launching hobbitd, I tried Henrik's debug command. The problem is I got no output after the command, bb hobbitserver-ip hobbitdboard Thanks for your pointer tj my hobbit server info 1. solaris sparc 10, 6/06, hobbit 4.2 2. errorr message from bbgen output Thu Sep 7 16:20:31 2006 bbgen for Hobbit version 4.2.0 Statistics: Hosts : 844 Status messages : 0 Purple messages : 0 Pages : 35 Error output: hobbitd status-board not available TIME SPENT Event Starttime Duration Startup 1157664031.590144 - Load links done 1157664031.590742 0.000598 Load bbhosts done 1157664031.618794 0.028052 ACK removal done 1157664031.618931 0.000137 Load STATE done 1157664031.620165 0.001234 Color calculation done 1157664031.620621 0.000456 Hobbit pagegen start 1157664031.620646 0.000025 Hobbit pagegen done 1157664032.090055 0.469409 BB2 generation done 1157664032.112544 0.022489 BBNK generation done 1157664032.121864 0.009320 Summary transmission done 1157664032.121867 0.000003 Run completed 1157664032.121868 0.000001 TIME TOTAL 0.531724 T.J. Yang
list Mike Shackelford
Hi, This is probably a dumb question, but how does hobbit server show alert if the hobbitlaunch process has died or become hung. I killed it as a test, and it doesnt seem to get data until hobbitlaunch is restarted. I would imagine that the hobbit server would need to interogate the client periodically beyond ping to ensure the main client deamon is running ok. Any help appreciated. thx. This is my first time using this type of tool, and it look great overall. -mike Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1¢/min.
list Allan Spencer
▸
Mike Shackelford wrote:
Hi, This is probably a dumb question, but how does hobbit server show alert if the hobbitlaunch process has died or become hung. I killed it as a test, and it doesnt seem to get data until hobbitlaunch is restarted. I would imagine that the hobbit server would need to interogate the client periodically beyond ping to ensure the main client deamon is running ok. Any help appreciated. thx. This is my first time using this type of tool, and it look great overall. -mike
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1¢/min. <http://us.rd.yahoo.com/mail_us/taglines/postman7/*http://us.rd.yahoo.com/evt=39666/*http://messenger.yahoo.com>;
The hobbit server will trigger a purple status for the client tests such as cpu disk etc if no events have been recieved for a period of time Cannot remember what the default time is for this but it can certainly be changed Allan
list Ralph Mitchell
▸
On 9/7/06, Allan Spencer <user-42a3456c44ef@xymon.invalid> wrote:
Mike Shackelford wrote:Hi, This is probably a dumb question, but how does hobbit server show alert if the hobbitlaunch process has died or become hung. I killed it as a test, and it doesnt seem to get data until hobbitlaunch is restarted. I would imagine that the hobbit server would need to interogate the client periodically beyond ping to ensure the main client deamon is running ok. Any help appreciated. thx. This is my first time using this type of tool, and it look great overall. -mike Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1¢/min. <http://us.rd.yahoo.com/mail_us/taglines/postman7/*http://us.rd.yahoo.com/evt=39666/*http://messenger.yahoo.com>;The hobbit server will trigger a purple status for the client tests such as cpu disk etc if no events have been recieved for a period of time Cannot remember what the default time is for this but it can certainly be changed
The default is to flag purple if no data is received in 30 minutes. Ralph Mitchell
list Mike Shackelford
Ok, I'm pretty sure I waited over 30 minutes and nothing changed, but I will check again tommorrow. Thanks for the feedback. -mike
▸
Ralph Mitchell <user-00a5e44c48c0@xymon.invalid> wrote:
The default is to flag purple if no data is received in 30 minutes.
Ralph Mitchell
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1¢/min.
list Henrik Størner
▸
On Thu, Sep 07, 2006 at 11:13:09AM -0500, Scheblein, Adam wrote:
2) When I put things in the "graphs" section it only changes the trends page. The whole reason that I originally made the change was so that when I click into the cpu info page, I would get the vmstat graph, however, I still wanted la and clock to continue graphing on the trends page. When I change CPU back to la both la and clock start graphing again, so I know the db's/rrd's are not corrupt.
Your method of changing it is incorrect. Right now you cannot get the vmstat graph on the "cpu" status display. Henrik
list Adam Scheblein
While you say i cannot or that it is incorrect, when i go to my cpu page, it is there. same with memory -- I have also replaced the memory graph (by changing the part in TEST2RRD to be memory=vmstat8) but like with CPU, now the memory graph has stopped graphing. Why do you say that the method is incorrect?? is there any better/cleaner way to do it?? I know you have said over and over again that it cannot be done, but i have done it and it seems to mostly work. Give it a shot and see that it does work, but has some weird side effects. (mainly clock and la stops graphing for cpu and memory stops graphing if you replace it with vmstat8 (or any other graph)) Adam
▸
From: Henrik Stoerner
Sent: Friday, September 08, 2006 4:00 AM
To: user-ae9b8668bcde@xymon.invalid
Subject: Re: [hobbit] 2 graphing questions
On Thu, Sep 07, 2006 at 11:13:09AM -0500, Scheblein, Adam wrote:2) When I put things in the "graphs" section it only changes the trends page. The whole reason that I originally made the change was so that when I click into the cpu info page, I would get the vmstat graph, however, I still wanted la and clock to continue graphing on the trends page. When I change CPU back to la both la and clock start graphing again, so I know the db's/rrd's are not corrupt.
Your method of changing it is incorrect. Right now you cannot get the vmstat graph on the "cpu" status display. Henrik
list Henrik Størner
When I said "incorrect" it was because of the side effects. It obviously serves your needs OK, so if you can live with the cpu- and memory-graphs not being updated, that is fine with me. Regards, Henrik
▸
On Fri, Sep 08, 2006 at 07:34:49AM -0500, Scheblein, Adam wrote:While you say i cannot or that it is incorrect, when i go to my cpu page, it is there. same with memory -- I have also replaced the memory graph (by changing the part in TEST2RRD to be memory=vmstat8) but like with CPU, now the memory graph has stopped graphing. Why do you say that the method is incorrect?? is there any better/cleaner way to do it?? I know you have said over and over again that it cannot be done, but i have done it and it seems to mostly work. Give it a shot and see that it does work, but has some weird side effects. (mainly clock and la stops graphing for cpu and memory stops graphing if you replace it with vmstat8 (or any other graph)) Adam From: Henrik Stoerner Sent: Friday, September 08, 2006 4:00 AM To: user-ae9b8668bcde@xymon.invalid Subject: Re: [hobbit] 2 graphing questions On Thu, Sep 07, 2006 at 11:13:09AM -0500, Scheblein, Adam wrote:2) When I put things in the "graphs" section it only changes the trends page. The whole reason that I originally made the change was so that when I click into the cpu info page, I would get the vmstat graph, however, I still wanted la and clock to continue graphing on the trends page. When I change CPU back to la both la and clock start graphing again, so I know the db's/rrd's are not corrupt.Your method of changing it is incorrect. Right now you cannot get the vmstat graph on the "cpu" status display. Henrik
--
Henrik Storner
list Rolf Schrittenlocher
Hi, running still hobbit-4.1.2p1 I noted that a conn test (ping) to a non-existing host gives back the correct message "no such host", but the colour is green. I'd have expected "red" regards Rolf
list Mike Shackelford
Thank you, I see purple this morning, I wasn't patient enough. I need to find the in's and out's of these config files to better understand it. -mike
▸
Ralph Mitchell <user-00a5e44c48c0@xymon.invalid> wrote: On 9/7/06, Allan Spencer wrote:Mike Shackelford wrote:Hi, This is probably a dumb question, but how does hobbit server show alert if the hobbitlaunch process has died or become hung. I killed it as a test, and it doesnt seem to get data until hobbitlaunch is restarted. I would imagine that the hobbit server would need to interogate the client periodically beyond ping to ensure the main client deamon is running ok. Any help appreciated. thx. This is my first time using this type of tool, and it look great overall. -mike Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1¢/min.The hobbit server will trigger a purple status for the client tests such as cpu disk etc if no events have been recieved for a period of time Cannot remember what the default time is for this but it can certainly be changed
The default is to flag purple if no data is received in 30 minutes. Ralph Mitchell
Get your own web address for just $1.99/1st yr. We'll help. Yahoo! Small Business.
list Adam Scheblein
Do you know why they stop graphing?? Thanks, Adam
▸
-----Original Message----- From: Henrik Stoerner [mailto:user-ce4a2c883f75@xymon.invalid] Sent: Friday, September 08, 2006 7:45 AM To: user-ae9b8668bcde@xymon.invalid Subject: Re: [hobbit] 2 graphing questions When I said "incorrect" it was because of the side effects. It obviously serves your needs OK, so if you can live with the cpu- and memory-graphs not being updated, that is fine with me. Regards, Henrik On Fri, Sep 08, 2006 at 07:34:49AM -0500, Scheblein, Adam wrote:While you say i cannot or that it is incorrect, when i go to my cpu page, it is there. same with memory -- I have also replaced the memory graph (by changing the part in TEST2RRD to be memory=vmstat8) but like with CPU, now the memory graph has stopped graphing. Why do you say that the method is incorrect?? is there any better/cleaner way to do it?? I know you have said over and over again that it cannot be done, but i have done it and it seems to mostly work. Give it a shot and see that it does work, but has some weird side effects. (mainly clock and la stops graphing for cpu and memory stops graphing if you replace it with vmstat8 (or any other graph)) Adam From: Henrik Stoerner Sent: Friday, September 08, 2006 4:00 AM To: user-ae9b8668bcde@xymon.invalid Subject: Re: [hobbit] 2 graphing questions On Thu, Sep 07, 2006 at 11:13:09AM -0500, Scheblein, Adam wrote:2) When I put things in the "graphs" section it only changes the trends page. The whole reason that I originally made the change was so that when I click into the cpu info page, I would get the vmstat graph, however, I still wanted la and clock to continue graphing on the trends page. When I change CPU back to la both la and clock start graphing again, so I know the db's/rrd's are not corrupt.Your method of changing it is incorrect. Right now you cannot get the vmstat graph on the "cpu" status display. Henrik-- Henrik Storner
list T.J. Yang
From reading at hyperic web site, I see GUI that I wish Hobbit/BB could have. but the licencse key part (see R1). By anychance the BB/Hobbit users evaluated Hyperic ? Regards T.J. Yang References: 1. http://www.hyperic.com/downloads/ 2. http://sourceforge.net/project/showfiles.php?group_id=172556&package_id=197470
list Larry Barber
That's odd. If hobbitlaunch isn't running, then bbgen will never be executed and your web pages shouldn't update. Thanks, Larry Barber
▸
On 9/8/06, Mike Shackelford <user-d51401b6b6e5@xymon.invalid> wrote:Thank you, I see purple this morning, I wasn't patient enough. I need to find the in's and out's of these config files to better understand it. -mike *Ralph Mitchell <user-00a5e44c48c0@xymon.invalid>* wrote: On 9/7/06, Allan Spencer wrote:Mike Shackelford wrote:Hi, This is probably a dumb question, but how does hobbit server show alert if the hobbitlaunch process has died or become hung. I killed it as a test, and it doesnt seem to get data until hobbitlaunch is restarted. I would imagine that the hobbit server would need to interogate the client periodically beyond ping to ensure the main client deamon is running ok. Any help appreciated. thx. This is my first time using this type of tool, and it look great overall. -mike Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls. Great rates starting at 1¢/min.The hobbit server will trigger a purple status for the client tests such as cpu disk etc if no events have been recieved for a period of time Cannot remember what the default time is for this but it can certainly be changedThe default is to flag purple if no data is received in 30 minutes. Ralph Mitchell Get your own web address for just $1.99/1st yr. We'll help. Yahoo! Small
Business<http://us.rd.yahoo.com/evt=41244/*http://smallbusiness.yahoo.com/>;.
list Ralph Mitchell
▸
On 9/8/06, Larry Barber <user-6ef9c2864140@xymon.invalid> wrote:
That's odd. If hobbitlaunch isn't running, then bbgen will never be executed and your web pages shouldn't update.
I got the impression he was killing the client side, not the server, since he mentions: "hobbit server would need to interogate the client periodically beyond ping" in which case the server would definitely notice that the client hadn't reported in over 30 mins. I see what you're saying about hobbitlaunch on the server, though. If *that's* down, all bets are off. Ralph Mitchell
list Henrik Størner
▸
On Fri, Sep 08, 2006 at 08:50:02AM -0500, Scheblein, Adam wrote:
Do you know why they stop graphing??
It's a long story, but what it really boils down to is that there are some silly dependencies between the status column names, the TEST2RRD settings, and the filename of the RRD file being updated. It's a mess, arising from the old BB system which didn't have RRD graphs well integrated, and there is no reason to keep it anymore except that I need to implement something else :-) The real solution to this is to first split the RRD updates away from the status-column updates; right now, most RRD file updates happen as a result of a status column update (and the data for the RRD file are taken from the status column message text), instead of being triggered by some data that has been collected (e.g. a host load average, the response time of a network service) which is fed directly into the RRD update mechanism. (This also means that Hobbit first spends time collecting all of the data, then generates a text message for the status column, and *then* it spends a lot of effort getting the data *out* of the status message again to put them into the RRD file. Not very clever). Second, you should be able to "compose" a status message from various bits and pieces of data that Hobbit has. There is no law saying that a "cpu" status column should include the load average data graph. Perhaps it should include some vmstat data graphs instead, or a user-count graph. Or both. OK, this is getting way off-topic, but - yes, I know why it stops graphing, and I know what it would take to get it right. So it *will* happen at some point :-) Regards, Henrik
list Mike Shackelford
Yep, the kill was on the client side only... -mike
▸
Ralph Mitchell <user-00a5e44c48c0@xymon.invalid> wrote: On 9/8/06, Larry Barber wrote:That's odd. If hobbitlaunch isn't running, then bbgen will never be executed and your web pages shouldn't update.
I got the impression he was killing the client side, not the server, since he mentions: "hobbit server would need to interogate the client periodically beyond ping" in which case the server would definitely notice that the client hadn't reported in over 30 mins. I see what you're saying about hobbitlaunch on the server, though. If *that's* down, all bets are off. Ralph Mitchell
All-new Yahoo! Mail - Fire up a more powerful email and get things done faster.