[tor-relays] Crashed Relay

I was notified by Uptime Robot that relay 1 of 3 had been shut down. Unfortunately I found this out 12 hours after the fact.

guard flag is lost.

journalctl -u tor@default

Nov 01 01:03:18 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: I learned some more directory information, but not enough to build a circuit: We need more microdescrors: we have 6178/6618, and can only build 49% of likely paths. (We have 99% of guards bw, 50% of midpoint bw, and 98% of exit bw = 49% of path bw.)

ptors: we have 6178/6618, and can only build 49% of likely paths. (We have 99% of guards bw, 50% of midpoint bw, and 98% of exit bw = 49% of path bw.)

ived 4371.61 GB. I’ve received 7345604 connections on IPv4 and 0 on IPv6. I’ve made 677780 connections with IPv4 and 0 with IPv6.

Nov 01 06:47:26 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Interrupt: we have stopped accepting new connections, and will shut down in 30 seconds. Interrupt aga>

Nov 01 06:47:26 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Delaying directory fetches: We are hibernating or shutting down.

Nov 01 06:47:56 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Clean shutdown finished. Exiting.

Nov 01 06:48:04 ubuntu-s-1vcpu-2gb-nyc1-01 systemd[1]: tor@default.service: Succeeded.

Nov 01 06:48:04 ubuntu-s-1vcpu-2gb-nyc1-01 systemd[1]: Stopped Anonymizing overlay network for TCP.

This is NOT a first time for the above on this relay, more like a 7th or 8th.

CPU usage according to NYX is 90/95%. That has been the last three weeks or so.

Prior to that, average CPU usage was 30% or less.

Current BW/Burst is 3/4 MBs on torrc and nyx. Prior was BW/Burst 4/5

*Me - retired fixed income. Average billing for 3 Digital Ocean Droplets was $40. budget is good with that. Octobers billing - $87.50 I used 10 gb instead of 6Gb. Essentially a $40 overage. It’s almost as if, torrc and nyx are being ignored.

···

Sent with ProtonMail Secure Email.

If you upgraded tor as a package on your Linux OS recently, it may have overwritten your torrc with defaults? It’s a long shot!

···

I was notified by Uptime Robot that relay 1 of 3 had been shut down. Unfortunately I found this out 12 hours after the fact.

guard flag is lost.

journalctl -u tor@default

Nov 01 01:03:18 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: I learned some more directory information, but not enough to build a circuit: We need more microdescrors: we have 6178/6618, and can only build 49% of likely paths. (We have 99% of guards bw, 50% of midpoint bw, and 98% of exit bw = 49% of path bw.)

ptors: we have 6178/6618, and can only build 49% of likely paths. (We have 99% of guards bw, 50% of midpoint bw, and 98% of exit bw = 49% of path bw.)

ived 4371.61 GB. I’ve received 7345604 connections on IPv4 and 0 on IPv6. I’ve made 677780 connections with IPv4 and 0 with IPv6.

Nov 01 06:47:26 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Interrupt: we have stopped accepting new connections, and will shut down in 30 seconds. Interrupt aga>

Nov 01 06:47:26 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Delaying directory fetches: We are hibernating or shutting down.

Nov 01 06:47:56 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Clean shutdown finished. Exiting.

Nov 01 06:48:04 ubuntu-s-1vcpu-2gb-nyc1-01 systemd[1]: tor@default.service: Succeeded.

Nov 01 06:48:04 ubuntu-s-1vcpu-2gb-nyc1-01 systemd[1]: Stopped Anonymizing overlay network for TCP.

This is NOT a first time for the above on this relay, more like a 7th or 8th.

CPU usage according to NYX is 90/95%. That has been the last three weeks or so.

Prior to that, average CPU usage was 30% or less.

Current BW/Burst is 3/4 MBs on torrc and nyx. Prior was BW/Burst 4/5

*Me - retired fixed income. Average billing for 3 Digital Ocean Droplets was $40. budget is good with that. Octobers billing - $87.50 I used 10 gb instead of 6Gb. Essentially a $40 overage. It’s almost as if, torrc and nyx are being ignored.

Sent with ProtonMail Secure Email.

So this log indicate that your "tor" received a SIGINT and did a clean
shutdown.

Not sure why that happened but it happened... :S

David

···

On 02 Nov (18:20:31), sysmanager7 via tor-relays wrote:

I was notified by Uptime Robot that relay 1 of 3 had been shut down. Unfortunately I found this out 12 hours after the fact.
guard flag is lost.

journalctl -u tor@default

Nov 01 01:03:18 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: I learned some more directory information, but not enough to build a circuit: We need more microdescrors: we have 6178/6618, and can only build 49% of likely paths. (We have 99% of guards bw, 50% of midpoint bw, and 98% of exit bw = 49% of path bw.)
ptors: we have 6178/6618, and can only build 49% of likely paths. (We have 99% of guards bw, 50% of midpoint bw, and 98% of exit bw = 49% of path bw.)
ived 4371.61 GB. I've received 7345604 connections on IPv4 and 0 on IPv6. I've made 677780 connections with IPv4 and 0 with IPv6.
Nov 01 06:47:26 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Interrupt: we have stopped accepting new connections, and will shut down in 30 seconds. Interrupt aga>
Nov 01 06:47:26 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Delaying directory fetches: We are hibernating or shutting down.
Nov 01 06:47:56 ubuntu-s-1vcpu-2gb-nyc1-01 Tor[328387]: Clean shutdown finished. Exiting.
Nov 01 06:48:04 ubuntu-s-1vcpu-2gb-nyc1-01 systemd[1]: tor@default.service: Succeeded.
Nov 01 06:48:04 ubuntu-s-1vcpu-2gb-nyc1-01 systemd[1]: Stopped Anonymizing overlay network for TCP.

--
fsQzn9293ONauAEmTyBPCVFpFcCWrEEXYnVe8hcOgOo=