Try what was suggested by Peter and all of my nodes appeared to be correct as nothing came up as bad.
Any other thoughts on what could be causing such a high load on the server? -- Gregory A. Thomas Student Life Support Specialist University of Wisconsin-Parkside thom...@uwp.edu<applewebdata://A3E407BF-7A5D-4A6C-A391-CA612A60CB27/owa/redir.aspx?C=PJoLX1MXo0SU0MLM7GrPmwxJzaMkdtAIgi4jkK-AXpCwJ307G0bt2lvFPw4WGoqQ06Tt1qwrKAA.&URL=mailto%3athomasg%40uwp.edu> 262.595.2432 From: "Truax, Peter via PacketFence-users" <packetfence-users@lists.sourceforge.net> Reply-To: "packetfence-users@lists.sourceforge.net" <packetfence-users@lists.sourceforge.net> Date: Thursday, September 6, 2018 at 12:18 PM To: "packetfence-users@lists.sourceforge.net" <packetfence-users@lists.sourceforge.net> Cc: "Truax, Peter" <ptr...@stmartin.edu> Subject: Re: [PacketFence-users] fingerbank-coll High load Gregory, The fingerbank.conf file is in /usr/local/fingerbank/ Not sure about your performance problem. I will tell you about a similar performance problem we had that we fixed. We had done several upgrades (version 7.4 – 8.1) during which an entry in the database got corrupted. This caused the whole system to slow down while the server tried to process the corrupted data. We were able to narrow down which entry it was by searching the nodes in pf. Some searches would be fast (1-10 seconds). Others would take a minute or longer. We eventually figured out which entry was our problem entry. We tried the command, ./pfcmd node delete xx:xx:xx:xx:xx:xx, but it did not work. It gave an error of “location log still active” or something similar. To fix it, we used the following commands: mysql –u root –p <enter password> use pf; delete from locationlog where mac = 'xx:xx:xx:xx:xx:xx'; exit; We were then able to run usr/local/pf/bin/pfcmd node delete xx:xx:xx:xx:xx:xx successfully. After deleting that entry, the server suddenly became much faster. I hope this information is helpful. Peter Truax Network Administrator St. Martin’s University Lacey, WA From: Thomas, Gregory A via PacketFence-users <packetfence-users@lists.sourceforge.net> Sent: Wednesday, September 5, 2018 7:21 AM To: packetfence-users@lists.sourceforge.net Cc: Thomas, Gregory A <thom...@uwp.edu> Subject: [PacketFence-users] fingerbank-coll High load I just started the semester, so there are a ton of new devices on the network and I expect there to be more load then when I first implemented the upgrade. Yesterday when there were reports of some lag, I took a look at the load and it was hovering around 15. The server has 4 virtual processors with 2 cores and 12 GB of RAM and 100GB of storage. It is the only VM on the box. It is version 8.1 and pf_maint and yum update were run yesterday and the system was rebooted right after. I know the next questions: yes I have the fingerbank API key installed. However, when looking there is no fingerbank.conf in the conf folder. When I search for any part of the key with ‘grep –r’, it is not found. Any clues on where to look next or what I can do to reduce the load? -- Gregory A. Thomas Student Life Support Specialist University of Wisconsin-Parkside thom...@uwp.edu<file:////owa/redir.aspx%3fC=PJoLX1MXo0SU0MLM7GrPmwxJzaMkdtAIgi4jkK-AXpCwJ307G0bt2lvFPw4WGoqQ06Tt1qwrKAA.&URL=mailto%253athomasg%2540uwp.edu> 262.595.2432
_______________________________________________ PacketFence-users mailing list PacketFence-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/packetfence-users
