Search results

Second Q: both links are DAC 10G sfp+ cables. One is up, second is still down even after 30m. It was hours after last reboot (month ago), before down link changed to up. I reconnectedboth cables on server side yesterday (forgot on the switch side). Bad cable? Some dwarf?

Hi, server has ens1f* interfaces down after PVE reboot. Both interfaces are bonded and used via openvswitch-switch. The first half shows ens1f* down and one VM starting (tap10110). The second half shows both links still down and VM failed to start, so tap iface is removed. All bond/OVS...

osd_pool_default_min_size = 2 osd_pool_default_size = 3 Calculate. You have 4 nodes with osds and for example 1 pool. Host replication (i think it's default). You have 3 copies and need 2 copies to have cluster working. What will happen, when you shutdown specific nodes holding those 2...

Isn't tls1/1.1 disabled anyway? Because PVE7.2 is based on Deb11 and it's openssl.conf has set MinProtocol = TLSv1.2 Hmm, it looks, postfix doesn't honor openssl configuration. SOMEHOST:~$ openssl s_client -connect PMG:25 -tls1 CONNECTED(00000003) 140408873498256:error:1408F10B:SSL...

netdata (cloud)

What? Any link? Doesn't pvesm depends on v3?

Ah, so email = recipient, Adddress = blacklisted sender. So, just not much helpful UI.

Hi, we added email to Administration->User Blacklist: But email is still delivered: Sep 29 12:52:34 pmg-01 pmg-smtp-filter[1815039]: 401A7633578ECED94F: SA score=1/5 time=5.279 bayes=0.00 autolearn=no autolearn_force=no...

Hm, 100GBit/s switch...show me please switch with almost 1Tbps ports.

FYI - this problem is not only PVE side, because last night maintenance it hit (semirandomly) about 5% VMs.

Opteron 6276 is very old and slow performant. Use better HW. Problem solved.

Hypervisor with 1 disk is nonsense if you don't have multiple such hypervisors with network storage. Solution: 2x ssd raid (pve+vm/cs) + hdd raid as data/backup storage only or 2x ssd raid (pve) + 2x ssd raid (vm/ct) etc.

"ceph-mgr Zabbix module" via https://docs.ceph.com/en/latest/mgr/zabbix/

Drop 10Gbps rj45 and use sfp+ only (10,25Gbps variant, dac or fibre). Power consumption and latency are better. You can use extra 2x nvme disks per node as other data pool.

Netdata don't care about host role. You need tweak its trigger levels if defaults are out of the way.

Same on hp dl380p g8 intel cpu. Probably solved in kernel patch mentioned in https://github.com/openshift/okd/issues/1189.

Check discard/trim functions.

Upload needed packages from USB.

Proxmox don't support drbd. Ask Linbit.

Only disk migration? Use shared/network storage or something as zfs sync.