OrbStack crashes when internet connection drops briefly #2322

Activity

1. panoptican

   added

   t/bugSomething isn't workingSomething isn't working

   on Jan 30, 2026
2. paunadeu commented on Mar 18, 2026

   

   paunadeu

   on Mar 18, 2026

   More actions

   Happening here too.
   Disconnected mount popup is also appeaing
3. ericmeadows commented on Mar 31, 2026

   

   ericmeadows

   on Mar 31, 2026

   More actions

   Same!
4. kdrag0n

   added this to the v2.1.4 milestone on May 23, 2026
5. kdrag0n commented on May 23, 2026

   

   kdrag0n

   on May 23, 2026

   Member

   More actions

   Fixed for the next version.
6. kdrag0n

   closed this as completedon May 23, 2026
7. kdrag0n commented on Jun 4, 2026

   

   kdrag0n

   on Jun 4, 2026

   Member

   More actions

   Released in v2.2.0.
8. kaloona commented on Aug 30, 2026

   

   kaloona

   on Aug 30, 2026

   More actions

   I hit this on OrbStack v2.2.3 (Docker 29.4.0), so whatever landed for the v2.2.0 milestone doesn't fully fix it.

   Sequence from ~/.orbstack/log/vmgr.log last night:

   🌸 agent:docker | time="08-29 22:52:49" level=error msg="failed to add network" error="too many interfaces" network=watchtower_default
   ...
   🌸 agent:docker | time="08-30 02:33:51" level=error msg="failed to add network" error="too many interfaces" network=bridge
   🌸 agent:docker | time="08-30 02:33:51" level=error msg="failed to add network" error="too many interfaces" network=mcp-camoufox_mcp-camoufox-net
   🌸 agent:docker | time="08-30 02:33:51" level=error msg="failed to add network" error="too many interfaces" network=searxng_searxng_internal
   🌸 agent:docker | time="08-30 02:33:51" level=error msg="failed to add network" error="too many interfaces" network=sitepins_sitepins-net
   🌸 agent:docker | time="08-30 02:33:51" level=error msg="failed to add network" error="too many interfaces" network=watchtower_default
   

   (repeats identically every few seconds for ~30 min, across every network on the host, not just one)

   It escalated into a full VM crash:

   🌲 vmgr | time="08-30 03:06:31" level=info msg="stop requested" reason=9
   🌲 vmgr | time="08-30 03:06:31" level=error msg="VM force stop failed" error="force stop already requested"
   2026-08-30 03:06:33.499 OrbStack[96835:...] Daemon exited: I/O error
   

   After that the VM stayed in Stopped state until I manually ran orb start — several hours of downtime for every container on the host.

   A few things that might help narrow it down:

   • At the time this started, the host had ~70 network interfaces total (46 veth, 13 br-*) — nowhere near any real kernel limit, consistent with the other report here of it happening at just 14 interfaces. So tooManyInterfaces looks like a bug in the agent's own interface accounting, not an actual resource exhaustion.
   • I checked macOS's configd logs (via log show) around both error timestamps and found no interface/connectivity state change — no Wi-Fi drop, no DHCP renewal, nothing. So at least in this case it doesn't look like it needs an actual "internet connection drop" to trigger; it may be firing on its own periodic reconciliation.
   • No crash-looping containers, no VPN/gluetun container running here, restart counts were all 0 except one unrelated container with 1 restart — so it's not caused by container churn on my end either.

   Happy to share the full vmgr.log if useful.

   👍React with 👍1Reacted by ᒍ ᗩ ᑕ K
9. JackCA commented on Sep 11, 2026

   

   JackCA

   on Sep 11, 2026

   More actions

   Happening here as well