You are not logged in.
Hi! I'm operating my network through netctl (no NetworkManager or anything else), connecting to wifi:
Description='Automatically generated profile by wifi-menu'
Interface=wlp3s0
Connection=wireless
Security=wpa
ESSID=[SSID]
IP=dhcp
Key=[password]
I'm suffering from periodic connection loss. Every 30 mintutes, the network goes down, and then needs to start again and get new DHCP address (which takes time and kills WebRTC connections). It happens regularly in our office network (some kind of mesh, maybe Cisco?), and something similar also happens at home where I have a single AP (but that may be an unrelated issue, so I'm focusing on the work network).
This is what it looks like in the journal:
Oct 18 08:44:10 mtpad2 kernel: wlp3s0: disassociated from 24:16:1b:cb:79:0d (Reason: 252=<unknown>)
Oct 18 08:44:10 mtpad2 dhcpcd[2091884]: wlp3s0: carrier lost
Oct 18 08:44:10 mtpad2 avahi-daemon[4236]: Withdrawing address record for 10.187.2.2 on wlp3s0.
Oct 18 08:44:10 mtpad2 avahi-daemon[4236]: Leaving mDNS multicast group on interface wlp3s0.IPv4 with address 10.187.2.2.
Oct 18 08:44:11 mtpad2 dhcpcd[2091884]: wlp3s0: deleting route to 10.187.0.0/22
Oct 18 08:44:11 mtpad2 avahi-daemon[4236]: Interface wlp3s0.IPv4 no longer relevant for mDNS.
Oct 18 08:44:11 mtpad2 dhcpcd[2091884]: wlp3s0: deleting default route via 10.187.0.1
Oct 18 08:44:12 mtpad2 systemd-resolved[3589]: Switching to fallback DNS server 1.1.1.1#cloudflare-dns.com.
Oct 18 08:44:13 mtpad2 kernel: wlp3s0: authenticate with 24:16:1b:ca:75:2d (local address=1c:1b:b5:89:5f:71)
Oct 18 08:44:13 mtpad2 kernel: wlp3s0: send auth to 24:16:1b:ca:75:2d (try 1/3)
Oct 18 08:44:13 mtpad2 kernel: wlp3s0: authenticated
Oct 18 08:44:13 mtpad2 kernel: wlp3s0: associate with 24:16:1b:ca:75:2d (try 1/3)
Oct 18 08:44:13 mtpad2 kernel: wlp3s0: RX AssocResp from 24:16:1b:ca:75:2d (capab=0x1111 status=0 aid=6)
Oct 18 08:44:13 mtpad2 kernel: wlp3s0: associated
Oct 18 08:44:13 mtpad2 dhcpcd[2091884]: wlp3s0: carrier acquired
Oct 18 08:44:13 mtpad2 kernel: wlp3s0: Limiting TX power to 23 (23 - 0) dBm as advertised by 24:16:1b:ca:75:2d
Oct 18 08:44:13 mtpad2 dhcpcd[2091884]: wlp3s0: IAID b5:89:5f:71
Oct 18 08:44:14 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 08:44:14 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 08:44:14 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 08:44:14 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 08:44:15 mtpad2 dhcpcd[2091884]: wlp3s0: rebinding lease of 10.187.2.2
Oct 18 08:44:15 mtpad2 dhcpcd[2091884]: wlp3s0: probing address 10.187.2.2/22
Oct 18 08:44:18 mtpad2 root[2132539]: ACPI group/action undefined: button/right / RIGHT
Oct 18 08:44:19 mtpad2 root[2132542]: ACPI group/action undefined: button/right / RIGHT
Oct 18 08:44:19 mtpad2 dhcpcd[2091884]: wlp3s0: leased 10.187.2.2 for 86400 seconds
Oct 18 08:44:19 mtpad2 avahi-daemon[4236]: Joining mDNS multicast group on interface wlp3s0.IPv4 with address 10.187.2.2.
Oct 18 08:44:19 mtpad2 avahi-daemon[4236]: New relevant interface wlp3s0.IPv4 for mDNS.
Oct 18 08:44:19 mtpad2 avahi-daemon[4236]: Registering new address record for 10.187.2.2 on wlp3s0.IPv4.
Oct 18 08:44:19 mtpad2 dhcpcd[2091884]: wlp3s0: adding route to 10.187.0.0/22
Oct 18 08:44:19 mtpad2 dhcpcd[2091884]: wlp3s0: adding default route via 10.187.0.1
Oct 18 08:44:19 mtpad2 systemd-resolved[3589]: wlp3s0: Bus client set DNS server list to: 10.187.0.1
The same thing happens again exactly 30 minutes later:
Oct 18 09:14:14 mtpad2 kernel: wlp3s0: disassociated from 24:16:1b:ca:75:2d (Reason: 252=<unknown>)
Oct 18 09:14:14 mtpad2 dhcpcd[2091884]: wlp3s0: carrier lost
Oct 18 09:14:14 mtpad2 avahi-daemon[4236]: Withdrawing address record for 10.187.2.2 on wlp3s0.
Oct 18 09:14:14 mtpad2 avahi-daemon[4236]: Leaving mDNS multicast group on interface wlp3s0.IPv4 with address 10.187.2.2.
Oct 18 09:14:14 mtpad2 dhcpcd[2091884]: wlp3s0: deleting route to 10.187.0.0/22
Oct 18 09:14:14 mtpad2 avahi-daemon[4236]: Interface wlp3s0.IPv4 no longer relevant for mDNS.
Oct 18 09:14:14 mtpad2 dhcpcd[2091884]: wlp3s0: deleting default route via 10.187.0.1
Oct 18 09:14:15 mtpad2 systemd-resolved[3589]: Switching to fallback DNS server 1.1.1.1#cloudflare-dns.com.
Oct 18 09:14:16 mtpad2 kernel: wlp3s0: authenticate with 24:16:1b:cb:79:0d (local address=1c:1b:b5:89:5f:71)
Oct 18 09:14:16 mtpad2 kernel: wlp3s0: send auth to 24:16:1b:cb:79:0d (try 1/3)
Oct 18 09:14:16 mtpad2 kernel: wlp3s0: send auth to 24:16:1b:cb:79:0d (try 2/3)
Oct 18 09:14:16 mtpad2 kernel: wlp3s0: authenticated
Oct 18 09:14:16 mtpad2 kernel: wlp3s0: associate with 24:16:1b:cb:79:0d (try 1/3)
Oct 18 09:14:16 mtpad2 kernel: wlp3s0: RX AssocResp from 24:16:1b:cb:79:0d (capab=0x1111 status=0 aid=3)
Oct 18 09:14:16 mtpad2 kernel: wlp3s0: associated
Oct 18 09:14:16 mtpad2 dhcpcd[2091884]: wlp3s0: carrier acquired
Oct 18 09:14:16 mtpad2 kernel: wlp3s0: Limiting TX power to 30 (30 - 0) dBm as advertised by 24:16:1b:cb:79:0d
Oct 18 09:14:16 mtpad2 dhcpcd[2091884]: wlp3s0: IAID b5:89:5f:71
Oct 18 09:14:17 mtpad2 dhcpcd[2091884]: wlp3s0: rebinding lease of 10.187.2.2
Oct 18 09:14:17 mtpad2 root[2146465]: ACPI group/action undefined: cd/play / CDPLAY
Oct 18 09:14:17 mtpad2 dhcpcd[2091884]: wlp3s0: probing address 10.187.2.2/22
Oct 18 09:14:17 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 09:14:17 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 09:14:17 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 09:14:17 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 09:14:18 mtpad2 py3status[15475]: received stop_signal SIGTSTP
Oct 18 09:14:21 mtpad2 dhcpcd[2091884]: wlp3s0: leased 10.187.2.2 for 86400 seconds
Oct 18 09:14:21 mtpad2 avahi-daemon[4236]: Joining mDNS multicast group on interface wlp3s0.IPv4 with address 10.187.2.2.
Oct 18 09:14:21 mtpad2 avahi-daemon[4236]: New relevant interface wlp3s0.IPv4 for mDNS.
Oct 18 09:14:21 mtpad2 avahi-daemon[4236]: Registering new address record for 10.187.2.2 on wlp3s0.IPv4.
Oct 18 09:14:21 mtpad2 dhcpcd[2091884]: wlp3s0: adding route to 10.187.0.0/22
Oct 18 09:14:21 mtpad2 dhcpcd[2091884]: wlp3s0: adding default route via 10.187.0.1
Oct 18 09:14:21 mtpad2 systemd-resolved[3589]: wlp3s0: Bus client set DNS server list to: 10.187.0.1
And again:
Oct 18 09:44:17 mtpad2 kernel: wlp3s0: disassociated from 24:16:1b:cb:79:0d (Reason: 252=<unknown>)
...
Oct 18 09:44:20 mtpad2 kernel: wlp3s0: associate with 24:16:1b:ca:75:2d (try 1/3)
It seems to be switching between two APs back and forth (while the PC stays at my desk all the time), and the network stack cannot take that. Could anyone please give me some pointers on what I can do to stabilize the connection while roaming? I found some config options that may help when using `systemd-networkd` (IgnoreCarrierLoss), but nothing for netctl.
I checked that I'm not running any conflicting network services (NM etc), which is mentioned several times in similar issues:
$ systemctl list-units --type=service --state=running
acpid.service loaded active running ACPI event daemon
avahi-daemon.service loaded active running Avahi mDNS/DNS-SD Stack
colord.service loaded active running Manage, Install and Generate Color Profiles
containerd.service loaded active running containerd container runtime
cups.service loaded active running CUPS Scheduler
dbus-broker.service loaded active running D-Bus System Message Bus
dirmngr@etc-pacman.d-gnupg.service loaded active running GnuPG network certificate management daemon for /etc/pacman.d/gnupg
docker.service loaded active running Docker Application Container Engine
getty@tty1.service loaded active running Getty on tty1
gpg-agent@etc-pacman.d-gnupg.service loaded active running GnuPG cryptographic agent and passphrase cache for /etc/pacman.d/gnupg
polkit.service loaded active running Authorization Manager
postgresql.service loaded active running PostgreSQL database server
rtkit-daemon.service loaded active running RealtimeKit Scheduling Policy Service
systemd-journald.service loaded active running Journal Service
systemd-logind.service loaded active running User Login Management
systemd-resolved.service loaded active running Network Name Resolution
systemd-timesyncd.service loaded active running Network Time Synchronization
systemd-udevd.service loaded active running Rule-based Manager for Device Events and Files
udisks2.service loaded active running Disk Manager
upower.service loaded active running Daemon for power management
user@1000.service loaded active running User Manager for UID 1000
Offline
You can try advising netctl to only use one of the APs:
AP=24:16:1b:cb:79:0d
Offline
You can try advising netctl to only use one of the APs:
That would be an option, thanks! But I do also move around the office from time to time, and need the network to able to switch APs when appropriate.
Offline
But at least you would know if the roaming availability of a second AP triggers your problem.
Offline
Good point, I configured the AP, we'll see what happens in 30 minutes.
Offline
OK, it's still failing, even when using a single AP, and on the original schedule, even though I re-connected manually (but did not reboot):
Oct 18 10:14:22 last reconnect to AP:79:0d
Oct 18 10:38:00 [update netctl config, set AP:79:0d]
Oct 18 10:39:57 mtpad2 network[2181618]: Stopping network profile 'Applifting'...
...
Oct 18 10:40:05 mtpad2 network[2181799]: Starting network profile 'Applifting'...
Oct 18 10:40:08 mtpad2 kernel: wlp3s0: authenticate with 24:16:1b:cb:79:0d (local address=1c:1b:b5:89:5f:71)
Oct 18 10:40:08 mtpad2 kernel: wlp3s0: send auth to 24:16:1b:cb:79:0d (try 1/3)
Oct 18 10:40:09 mtpad2 kernel: wlp3s0: authenticated
Oct 18 10:40:09 mtpad2 kernel: wlp3s0: associate with 24:16:1b:cb:79:0d (try 1/3)
Oct 18 10:40:09 mtpad2 kernel: wlp3s0: RX AssocResp from 24:16:1b:cb:79:0d (capab=0x1111 status=0 aid=6)
Oct 18 10:40:09 mtpad2 kernel: wlp3s0: associated
Oct 18 10:40:09 mtpad2 kernel: wlp3s0: Limiting TX power to 30 (30 - 0) dBm as advertised by 24:16:1b:cb:79:0d
Oct 18 10:40:09 mtpad2 dhcpcd[2181957]: dhcpcd-10.1.0 starting
Oct 18 10:40:09 mtpad2 dhcpcd[2181960]: DUID 00:01:00:01:2c:24:67:d1:1c:1b:b5:89:5f:71
Oct 18 10:40:09 mtpad2 dhcpcd[2181960]: wlp3s0: connected to Access Point: Applifting
Oct 18 10:40:09 mtpad2 dhcpcd[2181960]: wlp3s0: IAID b5:89:5f:71
Oct 18 10:40:10 mtpad2 avahi-daemon[4236]: Joining mDNS multicast group on interface wlp3s0.IPv6 with address fe80::1e1b:b5ff:fe89:5f71.
Oct 18 10:40:10 mtpad2 avahi-daemon[4236]: New relevant interface wlp3s0.IPv6 for mDNS.
Oct 18 10:40:10 mtpad2 avahi-daemon[4236]: Registering new address record for fe80::1e1b:b5ff:fe89:5f71 on wlp3s0.*.
Oct 18 10:40:10 mtpad2 dhcpcd[2181960]: wlp3s0: rebinding lease of 10.187.2.2
Oct 18 10:40:10 mtpad2 dhcpcd[2181960]: wlp3s0: probing address 10.187.2.2/22
Oct 18 10:40:15 mtpad2 dhcpcd[2181960]: wlp3s0: leased 10.187.2.2 for 86400 seconds
Oct 18 10:40:15 mtpad2 avahi-daemon[4236]: Joining mDNS multicast group on interface wlp3s0.IPv4 with address 10.187.2.2.
Oct 18 10:40:15 mtpad2 avahi-daemon[4236]: New relevant interface wlp3s0.IPv4 for mDNS.
Oct 18 10:40:15 mtpad2 avahi-daemon[4236]: Registering new address record for 10.187.2.2 on wlp3s0.IPv4.
Oct 18 10:40:15 mtpad2 dhcpcd[2181960]: wlp3s0: adding route to 10.187.0.0/22
Oct 18 10:40:15 mtpad2 dhcpcd[2181960]: wlp3s0: adding default route via 10.187.0.1
Oct 18 10:40:15 mtpad2 network[2182004]: Dropped protocol specifier '.dhcp' from 'wlp3s0.dhcp'. Using 'wlp3s0' (ifindex=3).
Oct 18 10:40:15 mtpad2 network[2181799]: Started network profile 'Applifting'
And here we go again:
Oct 18 10:44:23 mtpad2 dhcpcd[2181960]: wlp3s0: carrier lost
Oct 18 10:44:23 mtpad2 kernel: wlp3s0: disassociated from 24:16:1b:cb:79:0d (Reason: 252=<unknown>)
Oct 18 10:44:23 mtpad2 avahi-daemon[4236]: Withdrawing address record for 10.187.2.2 on wlp3s0.
Oct 18 10:44:23 mtpad2 avahi-daemon[4236]: Leaving mDNS multicast group on interface wlp3s0.IPv4 with address 10.187.2.2.
Oct 18 10:44:23 mtpad2 dhcpcd[2181960]: wlp3s0: deleting route to 10.187.0.0/22
Oct 18 10:44:23 mtpad2 dhcpcd[2181960]: wlp3s0: deleting default route via 10.187.0.1
Oct 18 10:44:23 mtpad2 avahi-daemon[4236]: Interface wlp3s0.IPv4 no longer relevant for mDNS.
Oct 18 10:44:23 mtpad2 systemd-resolved[3589]: Switching to fallback DNS server 1.1.1.1#cloudflare-dns.com.
Oct 18 10:44:25 mtpad2 kernel: wlp3s0: authenticate with 24:16:1b:cb:79:0d (local address=1c:1b:b5:89:5f:71)
Oct 18 10:44:25 mtpad2 kernel: wlp3s0: send auth to 24:16:1b:cb:79:0d (try 1/3)
Oct 18 10:44:25 mtpad2 kernel: wlp3s0: authenticated
Oct 18 10:44:25 mtpad2 kernel: wlp3s0: associate with 24:16:1b:cb:79:0d (try 1/3)
Oct 18 10:44:25 mtpad2 kernel: wlp3s0: RX AssocResp from 24:16:1b:cb:79:0d (capab=0x1111 status=0 aid=6)
Oct 18 10:44:25 mtpad2 kernel: wlp3s0: associated
Oct 18 10:44:25 mtpad2 dhcpcd[2181960]: wlp3s0: carrier acquired
Oct 18 10:44:25 mtpad2 kernel: wlp3s0: Limiting TX power to 30 (30 - 0) dBm as advertised by 24:16:1b:cb:79:0d
Oct 18 10:44:25 mtpad2 dhcpcd[2181960]: wlp3s0: IAID b5:89:5f:71
Oct 18 10:44:26 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 10:44:26 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 10:44:26 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 10:44:26 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 10:44:26 mtpad2 dhcpcd[2181960]: wlp3s0: rebinding lease of 10.187.2.2
Oct 18 10:44:27 mtpad2 dhcpcd[2181960]: wlp3s0: probing address 10.187.2.2/22
Oct 18 10:44:27 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 10:44:27 mtpad2 rtkit-daemon[15377]: Supervising 0 threads of 0 processes of 0 users.
Oct 18 10:44:32 mtpad2 dhcpcd[2181960]: wlp3s0: leased 10.187.2.2 for 86400 seconds
Oct 18 10:44:32 mtpad2 avahi-daemon[4236]: Joining mDNS multicast group on interface wlp3s0.IPv4 with address 10.187.2.2.
Oct 18 10:44:32 mtpad2 dhcpcd[2181960]: wlp3s0: adding route to 10.187.0.0/22
Oct 18 10:44:32 mtpad2 avahi-daemon[4236]: New relevant interface wlp3s0.IPv4 for mDNS.
Oct 18 10:44:32 mtpad2 avahi-daemon[4236]: Registering new address record for 10.187.2.2 on wlp3s0.IPv4.
Oct 18 10:44:32 mtpad2 dhcpcd[2181960]: wlp3s0: adding default route via 10.187.0.1
Oct 18 10:44:32 mtpad2 systemd-resolved[3589]: wlp3s0: Bus client set DNS server list to: 10.187.0.1
And again:
Oct 18 11:14:26 mtpad2 dhcpcd[2181960]: wlp3s0: carrier lost
Oct 18 11:14:26 mtpad2 kernel: wlp3s0: disassociated from 24:16:1b:cb:79:0d (Reason: 252=<unknown>)
...
So it may not have to do with the AP switching, and be the same problem as the one happening with my single AP home router.
Offline
Oct 18 10:40:15 mtpad2 dhcpcd[2181960]: wlp3s0: leased 10.187.2.2 for 86400 seconds
You're getting a 24h lease but that's probably a lie.
Try to set https://man.archlinux.org/man/extra/dhc … #leasetime 1680 (28 minutes)
Sanity check, there's no parallel windows installation, is there?
Offline
Now that it seems not to related to multiple APs it may be related to the access point model or a client problem.
Can you reproduce the error @home?
Offline
Connection at home seems to be stable. I configured the shorter DHCP leases, we'll see if it helps. I'll get back to the office and test it on monday. Thanks for the pointers!
Offline
Back to the office on monday, and back to the old schedule. The wifi keeps re-connecting every 30 minutes since I first logged into the network, even though I tried to reboot (midway through the interval) - which means that it is probably governed by the network rather than my system, as it syncs back to the initial DHCP address assignment.
But even though I configured the DHCP lease duration, and dhcpcd seems to have picked up the config, (Oct 21 08:45:55 mtpad2 dhcpcd[36791]: wlp3s0: leased 10.187.2.3 for 1680 seconds)) I don't see any attempts to renew the lease 28 minutes later. Should dhcpcd log something when it tries to renew the lease? (I'll also ask the network admins at the office if they see any irregularities on their side)
Offline
I found some information on Cisco Access Points having this interval set as the default value in a configurable option. Maybe it's the APs configuration.
Last edited by -thc (2024-10-21 08:06:13)
Offline
https://man.archlinux.org/man/extra/dhc … 5.en#debug - though I was under the impression dhcpcd would log lease renewals regardless.
Also try to add "wpa_ptk_rekey=1680" to the wpa_supplicant config.
Offline
Apparently, my laptop is the only device in the network that gets kicked out... The issue is probably at the wireless controller side rather then in my system configuration. Or at least something very specific is wrong with my configuration. I'll keep digging. Thanks for your help!
Offline
Did you talk to the IT?
Did they confirm the 24h lease time?
Did they not have logs as to why you get kicked?
Does dhcpcd still not log re-leases?
Also
Sanity check, there's no parallel windows installation, is there?
Offline
Yes, my IT colleagues told me that the wireless controller says:
.Oct 21 09:26:35.438: %CLIENT_EXCLUSION_SERVER-5-ADD_TO_EXCLUSIONLIST_REASON_DYNAMIC: Chassis 1 R0/0: wncmgrd: Client MAC: fe37.e8c7.22d5 was added to exclusion list associated with AP Name:AP-OpenSpace, BSSID:MAC: 2416.xxxx.xxxx, reason:VLAN failure
It's probably something to do with some network policy in the wireless controller which I know nothing about and can't affect. I'll have to investigate the network side more, I don't have much experience with this networking gear.
(DHCP lease renewal is logged on debug level at proper times, without any problems. It does not seem to have anything to do with the issue)
Offline
Offline