I have a 3 node docker based freeipa setup running with the rocky linux image. Starting from this weekend I can't enter the ui on one node. The other nodes work fine and replication runs.
Unsure. The setup is pretty basic.
I open the login interface and enter my credentials. After clicking "Log in" it show the message "Login failed due to an unknown reason".
In the httpd error.log I see the following line:
[Mon Jun 26 14:22:38.278497 2023] [wsgi:error] [pid 1172:tid 1348] [remote 192.168.96.1:36072] ipa: INFO: 401 Unauthorized: HTTPSConnectionPool(host='freeipa1.example.de', port=443): Max retries exceeded with url: /ipa/session/cookie (Caused by SSLError(SSLCertVerificationError(1, '[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate (_ssl.c:1129)')))
successful login
$ rpm -q freeipa-server freeipa-client ipa-server ipa-client 389-ds-base pki-ca krb5-server
[root@freeipa1-int /]# rpm -q freeipa-server freeipa-client ipa-server ipa-client 389-ds-base pki-ca krb5-server package freeipa-server is not installed package freeipa-client is not installed ipa-server-4.10.1-7.el9_2.x86_64 ipa-client-4.10.1-7.el9_2.x86_64 389-ds-base-2.2.4-3.el9.x86_64 package pki-ca is not installed krb5-server-1.20.1-9.el9_2.x86_64
I've searched a bit on the internet and found that this could be a django issue (very unsure about that..). Others suggest running 'ipa-certupdate' or 'ipa-cert-fix', which I tried without success.
I've started the container with this cmdline:
docker run -ti -h freeipa1-int.example.de --name freeipa1 --restart unless-stopped --read-only \ --add-host freeipa1.example.de:195.201.170.103 --add-host freeipa2.example.de:65.109.114.248 --add-host freeipa3.example.de:94.23.160.183 \ --add-host freeipa1:195.201.170.103 --add-host freeipa2:65.109.114.248 --add-host freeipa3:94.23.160.183 \ --add-host freeipa1-int.example.de:192.168.96.1 --add-host freeipa2-int.example.de:192.168.96.2 --add-host freeipa3-int.example.de:192.168.96.3 \ --add-host freeipa1-int:192.168.96.1 --add-host freeipa2-int:192.168.96.2 --add-host freeipa3-int:192.168.96.3 \ -v /srv/freeipa/data:/data:Z \ -e PASSWORD=SqXJ1ygUZbUCEyFUFhJC \ --sysctl net.ipv6.conf.lo.disable_ipv6=0 \ --sysctl net.ipv6.conf.eth0.disable_ipv6=0 \ --sysctl net.ipv6.conf.all.disable_ipv6=0 \ -p 53:53/udp -p 53:53 -p 81:80 -p 88:88/udp -p 88:88 -p 123:123/udp -p 389:389 -p 444:443 -p 464:464/udp -p 464:464 -p 636:636 -p 7389:7389 -p 9443:9443 -p 9444:9444 -p 9445:9445 \ freeipa/freeipa-server:rocky-9 ipa-server-install -r IPA.EXAMPLE.DE --domain=ipa.example.de --hostname=freeipa1-int.example.de --netbios-name=IPA --setup-adtrust --ssh-trust-dns --no-dnssec-validation --setup-dns --setup-kra --auto-forwarders --auto-reverse --allow-zone-overlap --no-ntp --unattended
Installation went fine and it was already running a few weeks. I also updated 2 or 3 times to a newer release already.
Another log message in httpd/error.log is (probably unrelated):
[Mon Jun 26 14:51:56.620061 2023] [wsgi:alert] [pid 38401:tid 38401] (2)No such file or directory: mod_wsgi (pid=38401): Unable to change working directory to home directory '/var/lib/kdcproxy' for uid=288. [Mon Jun 26 14:51:56.620082 2023] [wsgi:alert] [pid 38401:tid 38401] mod_wsgi (pid=38401): Failure to configure the daemon process correctly and process left in unspecified state. Restarting daemon process after delay.
Does anyone have a clue whats happening here?
best regards Oli
Docker Version is 24.0.2 running on Debian Buster
Running FreeIPA on a container is tricky, and container configuration should follow a strict set of rules, or everything falls apart.
Take a look at https://github.com/freeipa/freeipa-container. They have some instructions that should be closely followed, with very little margin for customization.
alright, thank you. I think I'll switch to a non-docker installation. I had a few other problems with it, but about this I have no clue.
Metadata Update from @magenbrot: - Issue close_status updated to: wontfix - Issue status updated to: Closed (was: Open)