Container won't fully start (seems like a clock issue?) with workaround

[zeddski]

After running for a year or two, without problems, the container crashed on me. After some fruitless debugging, removed container and images, trying to start over with default config (from github) using docker-compose . Doesn't work. Errors remain.
Please see below (workaround below logs).

unms_1 | [s6-init] making user provided files available at /var/run/s6/etc...exited 0.
unms_1 | [s6-init] ensuring user provided files have correct perms...exited 0.
unms_1 | [fix-attrs.d] applying ownership & permissions fixes...
unms_1 | [fix-attrs.d] done.
unms_1 | [cont-init.d] executing container initialization scripts...
unms_1 | [cont-init.d] 10-set-timezone: executing...
unms_1 | [cont-init.d] 10-set-timezone: exited 0.
unms_1 | [cont-init.d] 20-adduser: executing...
unms_1 |
unms_1 | -------------------------------------
unms_1 | GID/UID
unms_1 | -------------------------------------
unms_1 | User uid: 911
unms_1 | User gid: 911
unms_1 | -------------------------------------
unms_1 |
unms_1 | [cont-init.d] 20-adduser: exited 0.
unms_1 | [cont-init.d] 30-prepare: executing...
unms_1 | [cont-init.d] 30-prepare: exited 0.
unms_1 | [cont-init.d] 40-permissions: executing...
unms_1 | [cont-init.d] 40-permissions: exited 0.
unms_1 | [cont-init.d] 50-postgres: executing...
unms_1 | The files belonging to this database system will be owned by user "unms".
unms_1 | This user must also own the server process.
unms_1 |
unms_1 | The database cluster will be initialized with locales
unms_1 | COLLATE: C
unms_1 | CTYPE: C.UTF-8
unms_1 | MESSAGES: C
unms_1 | MONETARY: C
unms_1 | NUMERIC: C
unms_1 | TIME: C
unms_1 | The default database encoding has accordingly been set to "UTF8".
unms_1 | The default text search configuration will be set to "english".
unms_1 |
unms_1 | Data page checksums are disabled.
unms_1 |
unms_1 | creating directory /config/postgres ... ok
unms_1 | creating subdirectories ... ok
unms_1 | selecting dynamic shared memory implementation ... posix
unms_1 | selecting default max_connections ... 100
unms_1 | selecting default shared_buffers ... 128MB
unms_1 | selecting default time zone ... Australia/Sydney
unms_1 | creating configuration files ... ok
unms_1 | running bootstrap script ... ok
unms_1 | performing post-bootstrap initialization ... sh: locale: not found
unms_1 | 1970-05-01 16:13:04.010 AEST [327] WARNING: no usable system locales were found
unms_1 | ok
unms_1 | syncing data to disk ... ok
unms_1 |
unms_1 |
unms_1 | Success.
unms_1 |
unms_1 | initdb: warning: enabling "trust" authentication for local connections
unms_1 | You can change this by editing pg_hba.conf or using the option -A, or
unms_1 | --auth-local and --auth-host, the next time you run initdb.
unms_1 | [cont-init.d] 50-postgres: exited 0.
unms_1 | [cont-init.d] done.
unms_1 | [services.d] starting services
unms_1 | Starting rabbitmq-server...
unms_1 | Starting nginx...
unms_1 | Starting siridb-server...
unms_1 | Waiting for rabbitmq to start...
unms_1 | [services.d] done.
unms_1 | Running entrypoint.sh
unms_1 | /tmp:5432 - no response
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | Waiting for postgres to come up...
unms_1 | Starting postgres...
unms_1 | Creating nginx configuration
unms_1 | Enabling UNMS https and wss connections on port 443
unms_1 | 1970-05-06 05:26:40.010 AEST [426] LOG: starting PostgreSQL 13.5 on armv7-alpine-linux-musleabihf, compiled by gcc (Alpine 10.2.1_pre1) 10.2.1 20201203, 32-bit
unms_1 | 1970-05-06 05:26:40.010 AEST [426] LOG: listening on IPv4 address "127.0.0.1", port 5432
unms_1 | 1970-05-06 05:26:40.010 AEST [426] LOG: could not bind IPv6 address "::1": Address not available
unms_1 | 1970-05-06 05:26:40.010 AEST [426] HINT: Is another postmaster already running on port 5432? If not, wait a few seconds and retry.
unms_1 | 1970-05-06 05:26:40.010 AEST [426] LOG: listening on Unix socket "/run/postgresql/.s.PGSQL.5432"
unms_1 | 1970-05-06 05:26:40.010 AEST [426] LOG: listening on Unix socket "/tmp/.s.PGSQL.5432"
unms_1 | clock_gettimclock_gee(CLOCK_MONtOtTONIC, _)ime(CLOCK_M fONOTONICa, _) failied: Operaled: Operattiioonn nnoott ppeerrmmiitttteedd ((11))
unms_1 |
unms_1 | Generating self-signed certificate for 'localhost'.
unms_1 | Aborted (core dumped)
unms_1 | Aborted (core dumped)
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | 1970-05-06 05:26:40.010 AEST [588] FATAL: the database system is starting up
unms_1 | 1970-05-06 05:26:40.010 AEST [588] LOG: could not send data to client: Broken pipe
unms_1 | Starting rabbitmq-server...
unms_1 | clock_gettime(CLOCK_MONOTONIC, _) failed: Operation not permitted (1)
unms_1 | Aborted (core dumped)
unms_1 | Starting rabbitmq-server...
unms_1 | clock_gettime(CLOCK_MONOTONIC, _) failed: Operation not permitted (1)
unms_1 | Aborted (core dumped)
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | 1970-05-06 05:26:40.010 AEST [768] FATAL: the database system is starting up
unms_1 | 1970-05-06 05:26:40.010 AEST [768] LOG: could not send data to client: Broken pipe
unms_1 | 1970-05-06 05:26:40.010 AEST [770] FATAL: the database system is starting up
unms_1 | 1970-05-06 05:26:40.010 AEST [770] LOG: could not send data to client: Broken pipe
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | 1970-05-06 05:26:40.010 AEST [773] FATAL: the database system is starting up
unms_1 | 1970-05-06 05:26:40.010 AEST [773] LOG: could not send data to client: Broken pipe
unms_1 | Starting rabbitmq-server...
unms_1 | clock_gettime(CLOCK_MONOTONIC, _) failed: Operation not permitted (1)
unms_1 | 1970-05-06 05:26:40.010 AEST [426] LOG: startup process (PID 584) was terminated by signal 11: Segmentation fault
unms_1 | 1970-05-06 05:26:40.010 AEST [426] LOG: aborting startup due to startup process failure
unms_1 | 1970-05-06 05:26:40.010 AEST [426] LOG: database system is shut down
unms_1 | Aborted (core dumped)
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | Starting rabbitmq-server...
unms_1 | Starting postgres...
unms_1 | 1970-04-29 09:36:16.010 AEST [879] LOG: starting PostgreSQL 13.5 on armv7-alpine-linux-musleabihf, compiled by gcc (Alpine 10.2.1_pre1) 10.2.1 20201203, 32-bit
unms_1 | 1970-04-29 09:36:16.010 AEST [879] LOG: listening on IPv4 address "127.0.0.1", port 5432
unms_1 | 1970-04-29 09:36:16.010 AEST [879] LOG: could not bind IPv6 address "::1": Address not available
unms_1 | 1970-04-29 09:36:16.010 AEST [879] HINT: Is another postmaster already running on port 5432? If not, wait a few seconds and retry.
unms_1 | 1970-04-29 09:36:16.010 AEST [879] LOG: listening on Unix socket "/run/postgresql/.s.PGSQL.5432"
unms_1 | 1970-04-29 09:36:16.010 AEST [879] LOG: listening on Unix socket "/tmp/.s.PGSQL.5432"
unms_1 | clock_gettime(CLOCK_MONOTONIC, _) failed: Operation not permitted (1)
unms_1 | Aborted (core dumped)
unms_1 | Starting rabbitmq-server...
unms_1 | clock_gettime(CLOCK_MONOTONIC, _) failed: Operation not permitted (1)
unms_1 | Aborted (core dumped)
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | 1970-04-29 09:36:16.010 AEST [1053] FATAL: the database system is starting up
unms_1 | 1970-04-29 09:36:16.010 AEST [1053] LOG: could not send data to client: Broken pipe
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | 1970-04-29 09:36:16.010 AEST [1056] FATAL: the database system is starting up
unms_1 | 1970-04-29 09:36:16.010 AEST [1056] LOG: could not send data to client: Broken pipe
unms_1 | Starting rabbitmq-server...
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | 1970-04-29 09:36:16.010 AEST [1081] FATAL: the database system is starting up
unms_1 | 1970-04-29 09:36:16.010 AEST [1081] LOG: could not send data to client: Broken pipe
unms_1 | clock_gettime(CLOCK_MONOTONIC, _) failed: Operation not permitted (1)
unms_1 | Aborted (core dumped)
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | 1970-04-29 09:36:16.010 AEST [1150] FATAL: the database system is starting up
unms_1 | 1970-04-29 09:36:16.010 AEST [1150] LOG: could not send data to client: Broken pipe
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | 1970-04-29 09:36:16.010 AEST [1155] FATAL: the database system is starting up
unms_1 | 1970-04-29 09:36:16.010 AEST [1155] LOG: could not send data to client: Broken pipe
unms_1 | Starting rabbitmq-server...
unms_1 | 1970-04-29 09:36:16.010 AEST [879] LOG: startup process (PID 949) was terminated by signal 11: Segmentation fault
unms_1 | 1970-04-29 09:36:16.010 AEST [879] LOG: aborting startup due to startup process failure
unms_1 | 1970-04-29 09:36:16.010 AEST [879] LOG: database system is shut down
unms_1 | clock_gettime(CLOCK_MONOTONIC, _) failed: Operation not permitted (1)
unms_1 | /tmp:5432 - no response
unms_1 | Waiting for postgres to come up...
unms_1 | Aborted (core dumped)
unms_1 | Starting rabbitmq-server...
unms_1 | Starting postgres...

Ad Infinitum

What stands out:

• Clock has wrong time
• Hint of another process claiming port 5432. This is not the case.
• It seems a privileges-issue

Workaround: "privileged: true" in docker-compose configuration.
But this is too open, but I have no clue how to really fix this. Any clues?

[Nico640]

These clock errors sound like the same issue as in #59. Are you running this on a Raspberry Pi with a 32-Bit OS (Raspbian)?

If so, then this is most likely the libseccomp issue that occurs with an older libseccomp version (Debian Buster), 32bit ARM and Alpine Linux 3.13 which causes these time issues inside the container. Check if your libseccomp version is below 2.4.2. If it is, you'll need to update it. You can either update libseccomp manually, add buster-backports to your apt sources and update it that way, or update to Raspberry Pi OS 11 (Debian Bullseye), which has a newer libseccomp version.

This seems to be a pretty good explanation: https://docs.linuxserver.io/faq#libseccomp