Closed vroomr closed 4 years ago
Don't know where to look to fix.
Start with the Caddy log at /var/log/caddy.log--it should tell you why it's stopping.
On Thu, Dec 12, 2019 at 8:04 AM vroomr notifications@github.com wrote:
Have it with some regularity, and I do not know why.
Try to access my Nextcloud, and it is not running. I enter a shell, service caddy restart and everything is great.
Don't know where to look to fix. Tried adding caddy-restart.sh to crontab, but does not seem to help.
Currently on FreeNAS 11.3 Beta, with NC17.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/danb35/freenas-iocage-nextcloud/issues/78?email_source=notifications&email_token=AC4PNH6H7QDC4XLKTAN6BXTQYIZGTA5CNFSM4JZ6GWQKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IABOMSQ, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC4PNHYKMCG4NHFXUMUECUTQYIZGTANCNFSM4JZ6GWQA .
hmmmm. Currently not working. Last lines are:
2019/12/13 03:56:16 set tcp 192.XXX.XXX.XXX:XXXX->82.202.249.205:39251: setsockopt: connection reset by peer 2019/12/13 04:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 04:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 05:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 05:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 06:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 06:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 07:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 07:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 08:21:56 [INFO][cache:0xc0000d2a50] Scanning for expiring certificates 2019/12/13 08:21:56 [INFO][cache:0xc0000d2a50] Done scanning certificates 2019/12/13 08:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 08:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 09:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 09:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 10:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 10:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 11:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 11:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 12:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 12:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 13:21:56 [INFO][cache:0xc0000d2a50] Scanning for stale OCSP staples 2019/12/13 13:21:56 [INFO][cache:0xc0000d2a50] Done checking OCSP staples 2019/12/13 13:50:32 [INFO] SIGTERM: Shutting down servers then terminating 2019/12/13 13:50:32 [INFO][cache:0xc0000d2a50] Stopped certificate maintenance routine
Also, maybe noteworthy, maybe not, but there is lots of noise about stuff not looking like a TLS handshake.
2019/12/13 03:55:30 http: TLS handshake error from 76.65.215.21:48796: tls: first record does not look like a TLS handshake 2019/12/13 03:55:31 http: TLS handshake error from 93.115.28.171:63375: EOF 2019/12/13 03:55:36 http: TLS handshake error from 139.162.108.129:38768: EOF 2019/12/13 03:55:36 http: TLS handshake error from 139.162.108.129:44472: tls: first record does not look like a TLS handshake 2019/12/13 03:55:42 http: TLS handshake error from 139.162.108.129:44562: unexpected EOF 2019/12/13 03:55:42 http: TLS handshake error from 139.162.108.129:46506: tls: first record does not look like a TLS handshake 2019/12/13 03:55:42 http: TLS handshake error from 139.162.108.129:46690: tls: first record does not look like a TLS handshake 2019/12/13 03:55:48 http: TLS handshake error from 112.85.42.195:64418: tls: first record does not look like a TLS handshake
Any ideas?
Okey looking at this again we need a few things to be able to help and/or reproduce:
Tried installing with quite a few configurations... Never had this issue.
@vroomr I know yours issue:
2019/12/13 03:55:30 http: TLS handshake error from 76.65.215.21:48796: tls: first record does not look like a TLS handshake
Is the error you get when SNI mismatches. This happens for example when you put a reverse proxy in front of Nextcloud that doesn't forward SNI correctly.
Simply put: Caddy (by default) requires the SSL request to include a reference to the domain name, if it doesn't it freaks out. A fix for this is to set "default-sni" in the caddy config. Sadly enough @danb35 Rejected my proposal to include this. But that being said: This most likely your issue and/or your solution.
As this is quite old, you got a proposed solution now and we didn't hear anything, I think this can be closed... agree @danb35 ?
Thanks @Ornias1993 . Good with me to close it :)
Have it with some regularity, and I do not know why.
Try to access my Nextcloud, and it is not running. I enter a shell,
service caddy restart
and everything is great.Don't know where to look to fix. Tried adding caddy-restart.sh to crontab, but does not seem to help.
Currently on FreeNAS 11.3 Beta, with NC17.