Open-iscsi failed to start in two separate LXC containers on two different machines

So I updated my containers today. Ubuntu 16.04 base OS and container OS. Two of my containers, both “coincidentally” running an updated Apache2 instance, but both on completely different physical machines (with their LXD instance, own zpool etc.) gave an error as shown in the attached image. Now it might not be an issue with LXC, but just in case, I thought I’d post something here as a heads-up.

I can’t see anything other side effects yet, although “coincidentally”, my systems now seem very sluggish (but that could be my remote internet…).

Anyone have any idea as to why these errors would be triggered?



PS - am happy to post additional information if it helps track this down. I am not an expert, so please be specific (logs/file locations/names etc.). Thank you!

There’s really no good reason for open-iscsi to run in a container in the first place especially as the iscsi devices themselves aren’t namespaced at the kernel level.

I’d normally recommend removing that service or just ignoring it.

Looking at an Ubuntu 18.04 image here, I see open-iscsi installed, but it comes with startup conditions that appear to keep it from starting by default.

● open-iscsi.service - Login to default iSCSI targets
   Loaded: loaded (/lib/systemd/system/open-iscsi.service; enabled; vendor preset: enabled)
   Active: inactive (dead)
Condition: start condition failed at Mon 2018-11-05 17:06:31 UTC; 7min ago
           ├─ ConditionDirectoryNotEmpty=|/etc/iscsi/nodes was not met
           └─ ConditionDirectoryNotEmpty=|/sys/class/iscsi_session was not met
     Docs: man:iscsiadm(8)

Thank you. Glad it’s not a major issue. My container does seem to be working, but I wanted to report it in case it turned out to be significant. FYI I am running 16.04.

If I can just ignore it then I am fine with that. But if you think I should indeed ‘remove the service’ from my containers can you point me as to how I do that?