TMPFS getting full

Hi, i have an appliance with 8 GB RAM. i see a TMPFS of 4 GB.

in 3 days Tmpfs gets full and vpn, GUI, and some services stop working and i need to reboot.

I see these files in /tmp

It seems related to netifyd.

I’m using the latest patched version and this is a new fresh install

Thanks

Are there errors about netifyd in the logs?

Let’s check the netifyd status:

netifyd -s | grep -v UUID

A workaround is to reinstall the package:

opkg install --force-reinstall netifyd

We’re investigating the issue, could you send me privately the core dumps? I am not being able to reproduce it yet.

1 Like

root@firewall:~# netifyd -s | grep -v UUID
Netify Agent/5.1.25 (openwrt; linux-gnu; x86_64; conntrack; netlink; dns-cache; nfqueue; regex)
✓ agent is running: PID 14033
• agent timestamp: Wed Mar 4 12:49:58 2026
• agent uptime: 0d 00:00:45
! API updates: disabled
:left_arrow_curving_right: Netify API updates can be enabled from the configuration file:
/etc/netifyd.conf
! license manager available: no
✓ active flows: 737 / unlimited (0.0%)
• flows purged: 0, in-use: 97
• flows expiring: 0, expired: 0
• minimum flow size: 4672
• CPU cores: 4
✓ CPU utilization (user + system): 1.0%
✓ CPU time (user / system): 0.4s / 0.2s
• memory RSS: 28228 kB, maximum memory RSS: 28872 kB
✓ lan [LAN → NFQ]: online: packets dropped: 0.000%
✓ wan [WAN → NFQ]: online: packets dropped: 0.000%
• apps: 199, app-IP overrides: 0, domains: 3374, networks: 783, soft-dissectors: 26, transforms: 0
✓ DNS hint cache: enabled
• DHC entries: 1000, inserts: 1148 (10.4%), lookups: 613 (71.0%)
✓ flow hash cache: enabled
• FHC entries: 1000, inserts: 1136 (0.0%), lookups: 709 (1.3%)
• persistent state path: /etc/netifyd
• volatile state path: /var/run/netifyd
proc-aggregator/1.0.89 (processor)
• aggregator #1
• samples: 0
✓ license status: valid
proc-core/1.0.94 (processor)
proc-nfa/1.1.46 (processor)
✓ license status: valid
• actions: 1, enabled: 1
• targets: 3, enabled: 3
• global exemptions: 0
• flows: 266 / 266, queued: 1
sink-http/1.0.64 (sink)
sink-log/1.0.62 (sink)
sink-socket/1.0.76 (sink)

Thanks. Reinstalled, i have this output:

Collected errors:

  • resolve_conffiles: Existing conffile /etc/config/netifyd is different from the conffile in the new package. The new conffile will be placed at /etc/config/netifyd-opkg.

Worry not of the conffiles, it’s opkg saying that something changed in the config and won’t overwrite it. The update went good.

unfortunatly the tmpfs is slowly increasing again.

Rolled now an update for netifyd that avoids the fill, seems that the daemon crashes frequently on some systems, I’ve reached the developers in the meantime to figure out what is going on.

4 Likes

i’m updating. Thanks for the support!