https-dns-proxy: ubus command failed #22674
Assignees
Comments
systemcrash
changed the title
|
Looks like there's something else already running on port 5055. |
|
It does from the log messages. But I find nothing other than this daemon. Weird: nothing. Then it seems OK. Reboot router and I see the logs above anew. So the error is consistent. Possibly related to the |
|
At least on this boot I only got: the port in use things seemed absent: possibly owing to the time-change (ntp update)? in any case, a manual restart with |
|
Let me know if you need more info. |
|
Given that ubus error, I suspect it's caused by either this line: https://github.com/stangri/source.openwrt.melmac.net/blob/master/https-dns-proxy/files/etc/init.d/https-dns-proxy#L324 or explicit call here: https://github.com/stangri/source.openwrt.melmac.net/blob/master/https-dns-proxy/files/etc/init.d/https-dns-proxy#L139 I have no idea why it's not reproducible here tho. Are the logs above complete logs, so the service does not recover by itself at the end of the boot process? |
|
It appears as though the service recovers. DNS lookup succeed (also).
The logs above are complete. (As far as grep is concerned).
If you want me to test something specific, I can.
|
|
I have the same problem (device - Archer C7): |
|
@systemcrash @voron00 can you see if commenting out this line: https://github.com/stangri/source.openwrt.melmac.net/blob/master/https-dns-proxy/files/etc/init.d/https-dns-proxy#L138 and rebooting helps? |
Sadly no, with on_boot_trigger=1 removed behaviour is exactly the same. Here is a full log, maybe it will be more helpful: https://gist.github.com/voron00/5b24e20fd670ec57a7cc52fa7a97f835 |
The last message in that log is that service is stopped. Does it auto-recover/start later? |
Hmm, no it did not auto-recover. I have to manually restart it in order to get it working again. |
|
I can reintroduce the procd_boot_wait variable -- can you try inserting sleep command and try it with 5, 10, 20, 30 second wait, just below the https://github.com/stangri/source.openwrt.melmac.net/blob/master/https-dns-proxy/files/etc/init.d/https-dns-proxy#L138, above |
|
Adding sleep 5 or 30 does not solve the problem. |
|
@jow- any ideas why this: https://github.com/stangri/source.openwrt.melmac.net/blob/master/https-dns-proxy/files/etc/init.d/https-dns-proxy#L324 called from here: https://github.com/stangri/source.openwrt.melmac.net/blob/master/https-dns-proxy/files/etc/init.d/https-dns-proxy#L139 causes the error above on 23.05.2 and how can I fix that? |
|
Hello, The second variant succeeds: |
|
ubus -v list service
'service' @3fe5e78c
"set":{"name":"String","script":"String","instances":"Table","triggers":"Array","validate":"Array","autostart":"Boolean","data":"Table"}
"add":{"name":"String","script":"String","instances":"Table","triggers":"Array","validate":"Array","autostart":"Boolean","data":"Table"}
"list":{"name":"String","verbose":"Boolean"}
"delete":{"name":"String","instance":"String"}
"signal":{"name":"String","instance":"String","signal":"Integer"}
"update_start":{"name":"String"}
"update_complete":{"name":"String"}
"event":{"type":"String","data":"Table"}
"validate":{"package":"String","type":"String","service":"String"}
"get_data":{"name":"String","instance":"String","type":"String"}
"state":{"spawn":"Boolean","name":"String"}
"watchdog":{"mode":"Integer","timeout":"Integer","name":"String","instance":"String"} |
|
I think I have found a solution. Commenting out: helps and fixes the issue. I have inserted some debugging statements: and with them the output is:
There is a whole mess involved with calling causes the program to run well: |
Can you test if Without modifying the |
|
Yes, it does work. In syslog I got: The last message is "Stopping", but https-dns-proxy works and is used for DNS queries (which I verified by shutting it down manually - after it DNS queries stopped working). |
stangri
added a commit
to stangri/packages
that referenced
this issue
Nov 23, 2023
* fixes openwrt#22674 * rename resolver_health_check to is_resolver_running for readability Signed-off-by: Stan Grishin <stangri@melmac.ca>
stangri
added a commit
to stangri/packages
that referenced
this issue
Nov 23, 2023
* fixes openwrt#22674 * rename resolver_health_check to is_resolver_running for readability * reorder functions in the init file by name Signed-off-by: Stan Grishin <stangri@melmac.ca>
stangri
added a commit
to stangri/packages
that referenced
this issue
Nov 23, 2023
* fixes openwrt#22674 * rename resolver_health_check to is_resolver_running for readability * reorder functions in the init file by name Signed-off-by: Stan Grishin <stangri@melmac.ca> (cherry picked from commit f519b68)
stangri
added a commit
to stangri/packages
that referenced
this issue
Nov 23, 2023
* fixes openwrt#22674 * rename resolver_health_check to is_resolver_running for readability * reorder functions in the init file by name Signed-off-by: Stan Grishin <stangri@melmac.ca> (cherry picked from commit f519b68)
This was referenced Nov 23, 2023
tiagogaspar8
pushed a commit
to tiagogaspar8/packages
that referenced
this issue
Dec 20, 2023
* fixes openwrt#22674 * rename resolver_health_check to is_resolver_running for readability * reorder functions in the init file by name Signed-off-by: Stan Grishin <stangri@melmac.ca>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
@stangri ?
Just upgraded to 23.05.2,
logreadshows this:Here be my config:
The text was updated successfully, but these errors were encountered: