A strange behavior and an annoying warning are raised. When we look at the controllers, we see nodes with different versions, but the controllers change their version over time.
After a time
Hi there!
Could you provide more details (https://maas.io/docs/how-to-report-and-review-bugs) I’m not sure what the issue is?
Apologies for not providing much context earlier:
This is a new installation of MAAS, using the Debian installation method, with all machines running Ubuntu 22.04.
Here’s an overview of my environment (apologies for the ugly diagram):
+----------------------------------------------------+
| +--------------+ +--------------+ |
| | | | | |
| | Database -------------| RegionD | | +------------------------------------+
| | | | | | | +----------+ |
| +--------------+ +-------|------+ | | | | |
| | | | | Rackd | Physical Servers |
| | | | | | |
| +--------------+ | | +--|-------+ |
| | HA Proxy | | | | |
| | --------------------+ |
| +---|----------+ | | VLAN Y |
| VLAN X | | +------------------------------------+
+------------------------------------|---------------+
|
|
|
|
Users and/or Automations
After logging into MAAS, we see a warning:
In the “Controllers” section, we sometimes see rack controllers with the same version as the region, but other times they do not match (as shown in the image from the first post).
Upon checking the logs, I didn’t find any errors, just these messages:
Jul 26 11:31:24 maas-rackdd-2674-02 rackd[6543]: ClusterClient,client: [info] ClusterClient connection lost (HOST:IPv6Address(type='TCP', host='::ffff:10.107.74.21', port=43864, flowInfo=0, scopeID=0) PEER:IPv6Address(type='TCP', host='::ffff:10.107.71.20', port=5252, flowInfo=0, scopeID=0))
Jul 26 11:31:24 maas-rackdd-2674-02 rackd[6543]: ClusterClient,client: [info] ClusterClient connection lost (HOST:IPv6Address(type='TCP', host='::ffff:10.107.74.21', port=43878, flowInfo=0, scopeID=0) PEER:IPv6Address(type='TCP', host='::ffff:10.107.71.20', port=5252, flowInfo=0, scopeID=0))
In the regiond logs, we see the same thing:
Jul 26 11:31:24 maas-regiond01 regiond[7361]: RegionServer,2063,::ffff:10.107.74.21: [info] RegionServer connection lost (HOST:IPv6Address(type='TCP', host='::ffff:10.107.71.20', port=5252, flowInfo=0, scopeID=0) PEER:IPv6Address(type='TCP', host='::ffff:10.107.74.21', port=43864, flowInfo=0, scopeID=0))
Jul 26 11:31:24 maas-regiond01 regiond[7361]: RegionServer,2064,::ffff:10.107.74.21: [info] RegionServer connection lost (HOST:IPv6Address(type='TCP', host='::ffff:10.107.71.20', port=5252, flowInfo=0, scopeID=0) PEER:IPv6Address(type='TCP', host='::ffff:10.107.74.21', port=43878, flowInfo=0, scopeID=0))
Jul 26 11:31:24 maas-regiond01 regiond[7280]: maasserver.ipc: [info] Worker pid:7361 lost burst connection to ('10.107.71.20', 5252).
Jul 26 11:31:24 maas-regiond01 regiond[7280]: maasserver.ipc: [info] Worker pid:7361 lost burst connection to ('10.107.71.20', 5252).
I’m not sure what the issue is.
Hi @vasartori
It seems like you are hitting LP:2058063 bug
While the warning itself might be annoying, it should not affect overall MAAS stability