Search results

I'm experiencing this as well. The port disappears when the manager switches to another node. Now the active manager isn't binding 8003 for some reason. I tried disabling and enabling restful but to no avail. I have no idea how to debug this. Anyone experience this issue?

Yeah if I cancel a disk move to a ceph pool, it does say `Removing image: 1% complete...` but then is canceled at 2% so it seems that cancelling a disk move cancels the disk from being removed on the ceph pool. @Alwin

Derp I'm dumb, looks like I transferred a systemd network config file of /etc/systemd/network/99-default.link into the server with the contents: [Match] Path=/devices/virtual/net/* [Link] NamePolicy=kernel database onboard slot path MACAddressPolicy=none Removing that and rebuilding the...

There are no udev rules in /etc/udev/rules.d.

There was a new kernel update today/last night (9-19-2019) and it seems my servers' interface names reverted from ens0p0 to the old eth0 format. I don't have the predictable network interface names disabled in my grub startup parameters so why would it have reverted? Thanks!

Where did you read that KRBD cannot be used on the same machine that is running OSDs, since I'm having trouble finding information about that.

@Alwin did an update or two ago make changes to the calculation of available storage for ceph, as it's showing usage as expected for me now.

Sorry I meant it increases, not decreases haha. root@roc-server01:~# ceph versions { "mon": { "ceph version 14.2.1 (9257126ffb439de1652793b3e29f4c0b97a47b47) nautilus (stable)": 3 }, "mgr": { "ceph version 14.2.1 (9257126ffb439de1652793b3e29f4c0b97a47b47) nautilus...

Also when I move VMs off the ceph storage, the total amount decreases for some reason.

So my ceph pool reports its usage in the web gui as this: But then when going to the storage summary for the ceph pool it reports it as this: Is this normal or is there an issue going on that I need to resolve? Thanks!

I just removed the pool and rebuilt it and the issue was resolved. Thank you though!

When viewing the content on our ceph rbd within the Proxmox GUI, it displays this error message. No errors for running VMs on the ceph cluster or moving disks or anything. It's more annoying than anything but how can I resolve this issue without having to create a new pool and transfer data...

Thank you!

Same issue it's really annoying!

It's hard to compare but ~600mbps seems somewhat comparable, although I have less OSDs per node than the benchmarks, but what configuration changes were made in the benchmark to achieve the results?

I have setup a small ceph cluster with the following specs: Three Nodes identical - HP DL380p G8 - Intel Xeon E5-2697-v2 - 128GB DDR3 RAM (16GB 2RX4 PC3-14900R) - OS Drive: Intel DC S4500 240GB - OSD Drives: 2x Intel DC S3500 800GB - NIC: Intel X520 Configuration: [global] auth client required...