|
View:
New views
10 Messages
—
Rating Filter:
Alert me
|
|
|
Graphing ProblemFor some reason, there are gaps in my graphs. Here is a link to a copy:
http://img20.imageshack.us/img20/3139/cpudayhl1.png Anyone have any clue as to why? This system hasn't been down at all, yet there are gaps in the graph.u |
|
|
Re: Graphing ProblemMorfiusX wrote:
> For some reason, there are gaps in my graphs. Here is a link to a copy: > http://img20.imageshack.us/img20/3139/cpudayhl1.png > > Anyone have any clue as to why? This system hasn't been down at all, yet > there are gaps in the graph.u > Were you having cron problems? Did you get that taken care of? -Mike ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Efw-user mailing list Efw-user@... https://lists.sourceforge.net/lists/listinfo/efw-user |
|
|
Re: Graphing ProblemThis is a brand new installation. I'll check the logs to see if cron is dieing. From what I was reading though, cron doesn't get automatically restarted. It' just funny how it goes in ~30min increments of working/not working. Also, this is a VMware install, but I've never had this problem with other VMware installs.
Thanks MorfiusX
|
|
|
Re: Graphing ProblemHi
On Wed, 25 Oct 2006 04:50:00 -0700 (PDT), MorfiusX <morfiusx@...> wrote: > This is a brand new installation. I'll check the logs to see if cron is > dieing. From what I was reading though, cron doesn't get automatically i found that my cron is locking up from time to time... after some search i found that sometimes freshclam locks up when updating and the fcron doesnt continue until the process is killed As a workaround i edited the /etc/cron.hourly/freshclam and do a killall /usr/bin/freshclam && sleep 10 before running the real freshclam - "/usr/bin/freshclam &" i sent the process to background so it doesnt lock the fcron and i add the full path so the killall will kill the real freshclam and not this very same script, as it have the same name The endian people could add/fix this in their scripts as probably this appends to more people. > restarted. It' just funny how it goes in ~30min increments of > working/not working. Also, this is a VMware install, but I've never had but i'm betting that your problem is a clock problem... the exact time between holes point to it... virtual machines may have slower clocks than the real one because the way they use to account the time isnt stable in a virtual machine... in a 2.6 kernel there is a workaround of using clock=PIT or something like this (search the net) As the clock is slower, every time you run ntpdate in the cron it will jump in time to correct the clock, creating holes in the graphs. Other way to fix this is to run ntpd, this way the daemon will fix the clock in realtime and the jumps will stop while remaining with the correct clock. check it out good luck higuita -- Naturally the common people don't want war... but after all it is the leaders of a country who determine the policy, and it is always a simple matter to drag the people along, whether it is a democracy, or a fascist dictatorship, or a parliament, or a communist dictatorship. Voice or no voice, the people can always be brought to the bidding of the leaders. That is easy. All you have to do is tell them they are being attacked, and denounce the pacifists for lack of patriotism and exposing the country to danger. It works the same in every country. -- Hermann Goering, Nazi and war criminal, 1883-1946 ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Efw-user mailing list Efw-user@... https://lists.sourceforge.net/lists/listinfo/efw-user |
|
|
Re: Graphing Problemhi
Daniel Mota Leite (Higuita) wrote: > i found that my cron is locking up from time to time... > after some search i found that sometimes freshclam locks up when updating > and the fcron doesnt continue until the process is killed uhm. this may be a general problem, if cronscripts may block, since run-parts blocks on each script and starts the next script only if the actual exits. the other ones will be delayed.. if freshclam hangs, all scripts will be delayed. and on the next cron-turn it will not be started anymore since it is already running. :/ however, i tested with vixie cron on ubuntu, it's the same behaviour. i assume it's normal. tought about putting each cron script in background, but that maybe let us run into other problems (?) > As a workaround i edited the /etc/cron.hourly/freshclam and > do a killall /usr/bin/freshclam && sleep 10 before running the > real freshclam - "/usr/bin/freshclam &" cool! this should help. thank you again but maybe now it's time to let freshclam run as daemon (?) also maybe there is need to check the other cronscripts, maybe there are similar problems. > Other way to fix this is to run ntpd, this way the daemon will > fix the clock in realtime and the jumps will stop while remaining with > the correct clock. next release will use ntpd instead of ntpdate in order to address the problem with jumping clocks. stay tuned, it's ready soon :) peter -- :: e n d i a n :: open source - open minds :: peter warasin :: http://www.endian.it :: peter@... ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Efw-user mailing list Efw-user@... https://lists.sourceforge.net/lists/listinfo/efw-user |
|
|
Re: Graphing ProblemPeter Warasin wrote:
> hi > > Daniel Mota Leite (Higuita) wrote: > >> As a workaround i edited the /etc/cron.hourly/freshclam and >> do a killall /usr/bin/freshclam && sleep 10 before running the >> real freshclam - "/usr/bin/freshclam &" >> > > cool! this should help. thank you again > > but maybe now it's time to let freshclam run as daemon (?) > > > > > also adds a random delay. Mailscanner does this just so all the server do not try to hit the update servers at the same time. -Mike ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Efw-user mailing list Efw-user@... https://lists.sourceforge.net/lists/listinfo/efw-user |
|
|
Re: Graphing ProblemPeter Warasin wrote:
> hi > > Daniel Mota Leite (Higuita) wrote: > >> i found that my cron is locking up from time to time... >> after some search i found that sometimes freshclam locks up when updating >> and the fcron doesnt continue until the process is killed >> > > uhm. this may be a general problem, if cronscripts may block, since > run-parts blocks on each script and starts the next script only if the > actual exits. the other ones will be delayed.. if freshclam hangs, all > scripts will be delayed. and on the next cron-turn it will not be > started anymore since it is already running. :/ > > however, i tested with vixie cron on ubuntu, it's the same behaviour. i > assume it's normal. tought about putting each cron script in background, > but that maybe let us run into other problems (?) > > I've never noticed this. Where is run-parts coming from? I notice on a Redhat System run-parts is a shell script with this comment right at the top # run-parts - concept taken from Debian # keep going when something fails set +e I've never had a problem with items in cron.* blocking other cronjobs. That obviously is a problem. Anyone want to try swapping the run-parts from debianutils for a stock RedHat shell script? -Mike ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Efw-user mailing list Efw-user@... https://lists.sourceforge.net/lists/listinfo/efw-user |
|
|
Re: Graphing ProblemHi
On Fri, 27 Oct 2006 03:37:58 +0200, Peter Warasin <peter@...> wrote: > assume it's normal. tought about putting each cron script in background, > but that maybe let us run into other problems (?) i dont think that is usual for scripts to lockup, in fact i only saw freshclam locking up (and dont really know why yet) I remember that when i started to use unix systems, i learned that one should to send to background more complex cronjobs just in case they took longer, but not lock But if a process locks up it will stay there until next reboot, i had a freshclam locked for 3 days, before i note it... so there should be a way to clean up old process, specially in a appliance. The killall is a easy and simple shot term solution. The long term solution is to find why freshclam locks... i'm trying to get a log of a locked freshclam to know why... i will report when and if i found anything. > but maybe now it's time to let freshclam run as daemon (?) yes, it might be a good solution, for very frequent updates the daemon is better, for longer the crontab is the better way... But the questions is really: "will the deamon not lock up and freezes the AV updates?" Is there any "hidden" option to turn on the daemon or its only by editing the scripts? > also maybe there is need to check the other cronscripts, maybe there are > similar problems. again, i only saw the freshclam locked, but i will try to check if there are any others. > next release will use ntpd instead of ntpdate in order to address the > problem with jumping clocks. yes, i totally agree with this one... and it should be able to sync the clients in the network (via the ntp dhcp field -rarely used, but its there, so why not use it- and inform the admin in the docs how to manually configure the workstations to sync then, ie: net time /setsntp:IP_or_dns_of_efw ) thanks for endian higuita -- Naturally the common people don't want war... but after all it is the leaders of a country who determine the policy, and it is always a simple matter to drag the people along, whether it is a democracy, or a fascist dictatorship, or a parliament, or a communist dictatorship. Voice or no voice, the people can always be brought to the bidding of the leaders. That is easy. All you have to do is tell them they are being attacked, and denounce the pacifists for lack of patriotism and exposing the country to danger. It works the same in every country. -- Hermann Goering, Nazi and war criminal, 1883-1946 ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Efw-user mailing list Efw-user@... https://lists.sourceforge.net/lists/listinfo/efw-user |
|
|
Re: [Efw-user] Graphing Problemhi
Daniel Mota Leite (Higuita) wrote: > Is there any "hidden" option to turn on the daemon or its only > by editing the scripts? you can start freshclam as daemon by starting it with -d i left it as cronscript, otherwise i need to change clamav.cgi to reflect that change. furthermore there is only the possibility to configure how often the daemon checks for updates and not when exactly. we always heard from our customers that they want to know when the updates happen. i followed your advice and kill freshclam before running it. thank you! besides.. the new freshclam can timeout connections so this problem should be solved, since i applied mikes changes regarding the clamav update. >> next release will use ntpd instead of ntpdate in order to address the >> problem with jumping clocks. > yes, i totally agree with this one... and it should be able to > sync the clients in the network (via the ntp dhcp field -rarely used, > but its there, so why not use it- and inform the admin in the docs how > to manually configure the workstations to sync then, ie: > net time /setsntp:IP_or_dns_of_efw ) uhm, thats new for me :) i don't use windows for years now. but definitively worth adding to the docs. (christian? *g*) *yawn* peter -- :: e n d i a n :: open source - open minds :: peter warasin :: http://www.endian.it :: peter@... ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Efw-user mailing list Efw-user@... https://lists.sourceforge.net/lists/listinfo/efw-user |
|
|
|
| Free embeddable forum powered by Nabble | Forum Help |