I don't wish to flood the list, but I have some additional information that I 
think may be relevant. 

I see that periodically, usually spaced by hours, the dhcp listener log is 
written to, when that occurs the PF admin/queue display shows that the number 
for the queue count "pfdhcplisten" will drop to zero, often from the thousands.

At that moment the outstanding task counter for pfdhcplisten also drops, but 
only to 56, this number seems to always be 56 greater than the queue count, but 
clearly must grow over time.

The expired task counter for the pfdhcplistener is over 37 million and 
increases slowly.

I have also seen the listener queue reach only hundreds before being processed.

I suspect unprocessed/expired and/or outstanding tasks might explain why I have 
some items that are clearly showing in the pf log, in the dhcp log on the dhcp 
server but only have MAC and location information in the Packetfence database.

Suspect what I'm seeing is abnormal and wanted to present the information to 
the more experienced group.

The log is empty, the old log rolled to a .gz file and the new log isn't being 
written to.
Restarting the dhcp listener service did redirect the logging but did not cause 
the queue to flush.

Should I be concerned about this behavior and what action(s) should I take?

Thank you,
Brian


-----Original Message-----
From: Cuttler, Brian R (HEALTH) 
Sent: Thursday, August 4, 2022 2:52 PM
To: packetfence-users@lists.sourceforge.net
Subject: FW: [PacketFence-users] fail conect with new SSID



-----Original Message-----
From: Cuttler, Brian R (HEALTH) 
Sent: Thursday, August 4, 2022 10:40 AM
To: Zammit, Ludovic <luza...@akamai.com>
Subject: RE: [PacketFence-users] fail conect with new SSID

Can I use udp-reflector to send to both?

We were running it and sending to management port 1443, I have started a second 
instance sending to 767 and on the PF server I see traffic to both ports using 
tcpdump.

And I do see that we did log some traffic in the pfdhcplistener.log but it is 
not updating much, certainly not at the rate that we see incoming traffic via 
tcpdump, I had thought it was not updating at all.

Which is the correct socket to forward to, or should I always send to both?

I do see MAC addresses for dhcp clients that have no IP in PF in the dhcp log, 
so I know the clients have been active in the last day or so and can currently 
ping the IP.

Does the log only update when the information is new, or for all traffic?

Tail of log reads
[root@pkfn7 logs]# tail -f   pfdhcplistener.log
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] Updating iplog and SSO for 
b0:a8:6e:4c:f5:c0 -> 10.50.172.27 (pf::dhcp::processor::handle_new_ip)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] DHCPREQUEST from 
a8:d0:e5:32:64:00 (10.50.172.25) (pf::dhcp::processor::parse_dhcp_request)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] The listener process is NOT 
on the same server as the DHCP server. (pf::dhcp::processor::pf_is_dhcp)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] Updating iplog and SSO for 
a8:d0:e5:32:64:00 -> 10.50.172.25 (pf::dhcp::processor::handle_new_ip)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] DHCPREQUEST from 
ac:4b:c8:07:df:00 (10.49.65.250) (pf::dhcp::processor::parse_dhcp_request)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] The listener process is NOT 
on the same server as the DHCP server. (pf::dhcp::processor::pf_is_dhcp)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] Updating iplog and SSO for 
ac:4b:c8:07:df:00 -> 10.49.65.250 (pf::dhcp::processor::handle_new_ip)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] DHCPREQUEST from 
b0:a8:6e:47:ee:c0 (10.50.172.28) (pf::dhcp::processor::parse_dhcp_request)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] The listener process is NOT 
on the same server as the DHCP server. (pf::dhcp::processor::pf_is_dhcp)
Aug 04 07:20:44 pfqueue(45717) INFO: [mac:unknown] Updating iplog and SSO for 
b0:a8:6e:47:ee:c0 -> 10.50.172.28 (pf::dhcp::processor::handle_new_ip)

Thank you,
Brian

-----Original Message-----
From: Zammit, Ludovic <luza...@akamai.com> 
Sent: Thursday, August 4, 2022 10:15 AM
To: Cuttler, Brian R (HEALTH) <brian.cutt...@health.ny.gov>
Cc: packetfence-users@lists.sourceforge.net
Subject: Re: [PacketFence-users] fail conect with new SSID

[You don't often get email from luza...@akamai.com. Learn why this is important 
at https://aka.ms/LearnAboutSenderIdentification ]

ATTENTION: This email came from an external source. Do not open attachments or 
click on links from unknown senders or unexpected emails.



_______________________________________________
PacketFence-users mailing list
PacketFence-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/packetfence-users

Reply via email to