Re: [PVE-User] Some features that I miss in the PVE WebUI
On 22/05/2020 16:10, Thomas Lamprecht wrote: Hi, On 5/22/20 4:03 PM, Frank Thommen wrote: Dear all, having worked with oVirt in the past there are some features that I am really missing in PVE in my daily work: a) a tabular overview over all virtual machines. This should/might also include some performance data and the description. See the attached partial screenshot from oVirt, where this is implemented quite nicely. This is /not/ to replace proper monitoring but to provide a quick overview over the PVE-based intrastructure Isn't this what Datacenter -> Search could be? Note that the list drops any attachments. Partially. It is still missing some data that I really like in the oVirt overview (e.g. comments/descriptions, IP and the thumbnailed timelines(!) for CPU, memory and network ecc.) I have uploaded the screenshot to https://pasteboard.co/J9B4cEc.png. a1) the possibility to provide the virtual machines and containers with a short, one-line description Why not use the VM/CT notes? Editable over VM/CT -> Summary panel? We do use these notes heavily, but unfortunately they don't seem to be searchable/usable for filtering. It would be nice to have a "Description" which would be a one-liner, like the description for the HA resources. b) the possibility to use keywords from the Notes field or the description (see a1 above) in the search box. Our hosts are all named -vm which forces us to keep a separate list for the mapping of services to hostnames Dominik has done some patches adding "tags", which then would be searchable. Some backend support is there, but we had some discussion about how to integrate them in the frontend. I think this will be picked up soonish and should provide what you seek in b) maybe also a1.. yes, such tags would come in very handy and could partially replace the description if they can be used for filtering. Frank cheers, Thomas ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] Some features that I miss in the PVE WebUI
Dear all, having worked with oVirt in the past there are some features that I am really missing in PVE in my daily work: a) a tabular overview over all virtual machines. This should/might also include some performance data and the description. See the attached partial screenshot from oVirt, where this is implemented quite nicely. This is /not/ to replace proper monitoring but to provide a quick overview over the PVE-based intrastructure a1) the possibility to provide the virtual machines and containers with a short, one-line description b) the possibility to use keywords from the Notes field or the description (see a1 above) in the search box. Our hosts are all named -vm which forces us to keep a separate list for the mapping of services to hostnames It would be great to see these features in some future PVE release. Cheers, Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] UIDs > 65535 not valid in container
On 17.03.20 09:33, Dietmar Maurer wrote: Does anyone have an assessment of the risk we would run? I still don't understand the security implications of the mapping of higher UIDs. However this is quickly becoming a major issue for us. The risk is that it is not supported by us. Thus, we do not test that and I do not know what problems this may trigger... ok. I will take the risk then, because w/o that mapping we cannot use the containers. Thanks Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] UIDs > 65535 not valid in container
Dear all, On 13.03.20 14:13, Frank Thommen wrote: On 3/12/20 7:58 PM, Frank Thommen wrote: On 3/12/20 5:57 PM, Dietmar Maurer wrote: I fear this might be a container-related issue but I don't understand it and I don't know if there is a solution or a workaround. Any help or hint is highly appreciated Yes, we only map 65535 IDs for a single container. We cannot allow the full range for security reasons. What is the security related impact of higher UIDs? This is kind of a showstopper for us, as we planned several such minimal services which all need to be able to map all existing UIDs in the AD. The idea was to move them away from heavy full VMs to more lightweight containers. Or the other way round: What are the risks if we change the hardcoded limits in /usr/share/perl5/PVE/LXC.pm? (apart from the fact, that we will have to port the changes after each update and upgrade) Does anyone have an assessment of the risk we would run? I still don't understand the security implications of the mapping of higher UIDs. However this is quickly becoming a major issue for us. Cheers Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] UIDs > 65535 not valid in container
On 3/12/20 7:58 PM, Frank Thommen wrote: On 3/12/20 5:57 PM, Dietmar Maurer wrote: I fear this might be a container-related issue but I don't understand it and I don't know if there is a solution or a workaround. Any help or hint is highly appreciated Yes, we only map 65535 IDs for a single container. We cannot allow the full range for security reasons. What is the security related impact of higher UIDs? This is kind of a showstopper for us, as we planned several such minimal services which all need to be able to map all existing UIDs in the AD. The idea was to move them away from heavy full VMs to more lightweight containers. Or the other way round: What are the risks if we change the hardcoded limits in /usr/share/perl5/PVE/LXC.pm? (apart from the fact, that we will have to port the changes after each update and upgrade) frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] UIDs > 65535 not valid in container
On 3/12/20 5:57 PM, Dietmar Maurer wrote: I fear this might be a container-related issue but I don't understand it and I don't know if there is a solution or a workaround. Any help or hint is highly appreciated Yes, we only map 65535 IDs for a single container. We cannot allow the full range for security reasons. What is the security related impact of higher UIDs? This is kind of a showstopper for us, as we planned several such minimal services which all need to be able to map all existing UIDs in the AD. The idea was to move them away from heavy full VMs to more lightweight containers. Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] UIDs > 65535 not valid in container
On 3/12/20 6:10 PM, Daniel Berteaud wrote: - Le 12 Mar 20, à 16:35, Frank Thommen f.thom...@dkfz-heidelberg.de a écrit : Dear all, we have a strange issue with a CentOS 7 container running on PVE 6.1-3, that UIDs > 65535 are invalid. The container is used as a "SSH jumphost" to access a special network: Users log in to the host and SSH to the special network from there. sssd is running in the container. The directory service is an Active Directory. However users with UID > 65535 cannot login: /var/log/secure: [...] Mar 12 13:48:32 XX sshd[1021]: fatal: seteuid 86544: Invalid argument [...] and chown isn't possible either: $ chown 65535 /home/test $ chown 65536 /home/test chown: changing ownership of ‘/home/test’: Invalid argument $ There are no problems with such UIDs on any other systems and there is no problem with users with an UID <= 65535 within the container. I fear this might be a container-related issue but I don't understand it and I don't know if there is a solution or a workaround. Any help or hint is highly appreciated You can work with higher UID in LXC with this : * Edit /etc/subuid and change the range. Eg root:10:400039 * Do the same for /etc/subgid * Edit your container config (/etc/pve/lxc/XXX.conf) and add lxc.idmap: u 0 10 200020 lxc.idmap: g 0 10 200020 That's the values I'm using for some AD members containers. Note however that native PVE restore code might refuse to work with those UID (I recall the 65535 max UID hardcoded somewhere in the restore path, but can't remember exactly where) Unfortunately that doesn't work. The container will not start any more with the following messages in the debug log (shortened): [...] lxc-start 101 20200312185335.631 INFO conf - conf.c:run_script_argv:372 - Executing script "/usr/share/lxc/hooks/lxc-pve-prestart-hook" for container "101", config section "lxc" lxc-start 101 20200312185336.964 DEBUGconf - conf.c:run_buffer:340 - Script exec /usr/share/lxc/hooks/lxc-pve-prestart-hook 101 lxc pre-start produced output: unable to detect OS distribution lxc-start: 101: conf.c: run_buffer: 352 Script exited with status 2 lxc-start: 101: start.c: lxc_init: 897 Failed to run lxc.hook.pre-start for container "101" lxc-start: 101: start.c: __lxc_start: 2032 Failed to initialize container "101" Segmentation fault Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] UIDs > 65535 not valid in container
Dear all, we have a strange issue with a CentOS 7 container running on PVE 6.1-3, that UIDs > 65535 are invalid. The container is used as a "SSH jumphost" to access a special network: Users log in to the host and SSH to the special network from there. sssd is running in the container. The directory service is an Active Directory. However users with UID > 65535 cannot login: /var/log/secure: [...] Mar 12 13:48:32 XX sshd[1021]: fatal: seteuid 86544: Invalid argument [...] and chown isn't possible either: $ chown 65535 /home/test $ chown 65536 /home/test chown: changing ownership of ‘/home/test’: Invalid argument $ There are no problems with such UIDs on any other systems and there is no problem with users with an UID <= 65535 within the container. I fear this might be a container-related issue but I don't understand it and I don't know if there is a solution or a workaround. Any help or hint is highly appreciated Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Why are images and rootdir not supported on CephFS?
On 2/17/20 7:23 AM, Fabian Grünbichler wrote: On February 16, 2020 5:26 pm, Frank Thommen wrote: Thank you for the link. Even though Fabian Gruenbichler writes in the bugreport (https://bugzilla.proxmox.com/show_bug.cgi?id=2490#c2) that Ceph RBD offers all features of CephFS, this doesn't seem to be true(?), as CephFS supports "vztmpl iso backup snippets" and Ceph RBD "images rootdir" (https://pve.proxmox.com/pve-docs/chapter-pvesm.html), so these two storage types are complementary and RBD cannot replace CephFS. that comment was about using CephFS for guest image/volume storage. for that use case, RBD has all the same features (even more), but with better performance. obviously I didn't mean that RBD is a file system ;) What would be a good practice (CephFS and RBD are already set up): Create an RBD storage on the same (PVE based) Ceph storage that already has CephFS on top of it and use one for templates and backups and the other for images and rootdir? yes Won't it create problems when using the same Ceph pool with CephFS /and/ RBD (this is probably rather a Ceph question, though) you don't use the same Ceph pool, just the same OSDs (pools are logical in Ceph, unlike with ZFS), so this is not a problem. Additionally this might create problems with our inhouse tape backup, as I don't think it supports backing up object storage... the usual backup options are available - use vzdump, and then backup the VMA files. or use some backup solution inside the guest. or both ;) Thank you for all the hints above. I will go along these lines. Still struggling with some Ceph concepts ;-) frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Why are images and rootdir not supported on CephFS?
I have now configured a Directory storage which points to the CephFS mountpoint. When creating a container (Alpine Linux, 10 GB disk, 2 GB Memory), this happens in the blink of an eye when using Ceph RBD or local storage as root disk, but it takes very, very long (10 to 20 times longer) when using the CephFS-directory as root disk. On 16/02/2020 17:26, Frank Thommen wrote: Thank you for the link. Even though Fabian Gruenbichler writes in the bugreport (https://bugzilla.proxmox.com/show_bug.cgi?id=2490#c2) that Ceph RBD offers all features of CephFS, this doesn't seem to be true(?), as CephFS supports "vztmpl iso backup snippets" and Ceph RBD "images rootdir" (https://pve.proxmox.com/pve-docs/chapter-pvesm.html), so these two storage types are complementary and RBD cannot replace CephFS. What would be a good practice (CephFS and RBD are already set up): Create an RBD storage on the same (PVE based) Ceph storage that already has CephFS on top of it and use one for templates and backups and the other for images and rootdir? Won't it create problems when using the same Ceph pool with CephFS /and/ RBD (this is probably rather a Ceph question, though) Additionally this might create problems with our inhouse tape backup, as I don't think it supports backing up object storage... frank On 15/02/2020 00:38, Gianni Milo wrote: This has been discussed in the past, see the post below for some answers... https://www.mail-archive.com/pve-user@pve.proxmox.com/msg10160.html On Fri, 14 Feb 2020 at 22:57, Frank Thommen wrote: Dear all, the PVE documentation on https://pve.proxmox.com/pve-docs/chapter-pvesm.html#_storage_types says, that "[File level based storage technologies] allow you to store content of any type". However I found now - after having combined all available disks in a big CephFS setup - that this is not true for CephFS, which does not support images and rootdir. What is the reason, that of all filesystens, CephFS doesn't support the main PVE content type? :-) Ceph RPD on the other hand, doesn't support backup. This seems to rule our Ceph (in whatever variant) as an unifying, shared storage for PVE. Or do I miss an important point here? Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Why are images and rootdir not supported on CephFS?
Thank you for the link. Even though Fabian Gruenbichler writes in the bugreport (https://bugzilla.proxmox.com/show_bug.cgi?id=2490#c2) that Ceph RBD offers all features of CephFS, this doesn't seem to be true(?), as CephFS supports "vztmpl iso backup snippets" and Ceph RBD "images rootdir" (https://pve.proxmox.com/pve-docs/chapter-pvesm.html), so these two storage types are complementary and RBD cannot replace CephFS. What would be a good practice (CephFS and RBD are already set up): Create an RBD storage on the same (PVE based) Ceph storage that already has CephFS on top of it and use one for templates and backups and the other for images and rootdir? Won't it create problems when using the same Ceph pool with CephFS /and/ RBD (this is probably rather a Ceph question, though) Additionally this might create problems with our inhouse tape backup, as I don't think it supports backing up object storage... frank On 15/02/2020 00:38, Gianni Milo wrote: This has been discussed in the past, see the post below for some answers... https://www.mail-archive.com/pve-user@pve.proxmox.com/msg10160.html On Fri, 14 Feb 2020 at 22:57, Frank Thommen wrote: Dear all, the PVE documentation on https://pve.proxmox.com/pve-docs/chapter-pvesm.html#_storage_types says, that "[File level based storage technologies] allow you to store content of any type". However I found now - after having combined all available disks in a big CephFS setup - that this is not true for CephFS, which does not support images and rootdir. What is the reason, that of all filesystens, CephFS doesn't support the main PVE content type? :-) Ceph RPD on the other hand, doesn't support backup. This seems to rule our Ceph (in whatever variant) as an unifying, shared storage for PVE. Or do I miss an important point here? Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] Why are images and rootdir not supported on CephFS?
Dear all, the PVE documentation on https://pve.proxmox.com/pve-docs/chapter-pvesm.html#_storage_types says, that "[File level based storage technologies] allow you to store content of any type". However I found now - after having combined all available disks in a big CephFS setup - that this is not true for CephFS, which does not support images and rootdir. What is the reason, that of all filesystens, CephFS doesn't support the main PVE content type? :-) Ceph RPD on the other hand, doesn't support backup. This seems to rule our Ceph (in whatever variant) as an unifying, shared storage for PVE. Or do I miss an important point here? Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Ceph: Monitors not running but cannot be destroyed or recreated
On 26/01/2020 16:46, Frank Thommen wrote: On 26/01/2020 14:14, Frank Thommen wrote: Dear all, I am trying to destroy "old" Ceph monitors but they can't be deleted and also cannot be recreated: I am currently configuring Ceph on our PVE cluster (3 nodes running PVE 6.1-3). There have been some "remainders" of a previous Ceph configuration which I had tried to configure while the nodes were not in a cluster configuration yet (and I had used the wrong network). However I had purged these configurations with `pveceph purge`. I have redone the basic Ceph configuration through the GUI on the first node and I have deleted the still existing managers through the GUI (to have a fresh start). A new monitor has been created on the first node automatically, but I am unable to delete the monitors on nodes 2 and 3. They show up as Status=stopped and Address=Unknown in the GUI and they cannot be started (no error message). In the syslog window I see (after rebooting node odcf-pve02): Jan 26 13:51:53 odcf-pve02 systemd[1]: Started Ceph cluster monitor daemon. Jan 26 13:51:55 odcf-pve02 ceph-mon[1372]: 2020-01-26 13:51:55.450 7faa98ab9280 -1 mon.odcf-pve02@0(electing) e1 failed to get devid for : fallback method has serial ''but no model On the other hand I see the same message on the first node, and there the monitor seems to work fine. Trying to destroy them results in the message, that there is no such monitor, and trying to create a new monitor on these nodes results in the message, that the monitor already exists I am stuck in this existence loop. Destroying or creating them also doesn't work on the commandline. Any idea on how to fix this? I'd rather not completely reinstall the nodes :-) Cheers frank In an attempt to clean up the Ceph setup again, I ran pveceph stop ceph.target pveceph purge on the first node. Now I get an rados_connect failed - No such file or directory (500) when I select Ceph in the GUI of any of the three nodes. A reboot of all nodes didn't help. frank I was finally able to completely purge the old settings and reconfigure Ceph with the various instructions from this (https://forum.proxmox.com/threads/not-able-to-use-pveceph-purge-to-completely-remove-ceph.59606/) post. Maybe this information could be added to the official documentation (unless there is a nicer way of completely resetting Ceph in a PROXMOX cluster)? frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Ceph: Monitors not running but cannot be destroyed or recreated
On 26/01/2020 14:14, Frank Thommen wrote: Dear all, I am trying to destroy "old" Ceph monitors but they can't be deleted and also cannot be recreated: I am currently configuring Ceph on our PVE cluster (3 nodes running PVE 6.1-3). There have been some "remainders" of a previous Ceph configuration which I had tried to configure while the nodes were not in a cluster configuration yet (and I had used the wrong network). However I had purged these configurations with `pveceph purge`. I have redone the basic Ceph configuration through the GUI on the first node and I have deleted the still existing managers through the GUI (to have a fresh start). A new monitor has been created on the first node automatically, but I am unable to delete the monitors on nodes 2 and 3. They show up as Status=stopped and Address=Unknown in the GUI and they cannot be started (no error message). In the syslog window I see (after rebooting node odcf-pve02): Jan 26 13:51:53 odcf-pve02 systemd[1]: Started Ceph cluster monitor daemon. Jan 26 13:51:55 odcf-pve02 ceph-mon[1372]: 2020-01-26 13:51:55.450 7faa98ab9280 -1 mon.odcf-pve02@0(electing) e1 failed to get devid for : fallback method has serial ''but no model On the other hand I see the same message on the first node, and there the monitor seems to work fine. Trying to destroy them results in the message, that there is no such monitor, and trying to create a new monitor on these nodes results in the message, that the monitor already exists I am stuck in this existence loop. Destroying or creating them also doesn't work on the commandline. Any idea on how to fix this? I'd rather not completely reinstall the nodes :-) Cheers frank In an attempt to clean up the Ceph setup again, I ran pveceph stop ceph.target pveceph purge on the first node. Now I get an rados_connect failed - No such file or directory (500) when I select Ceph in the GUI of any of the three nodes. A reboot of all nodes didn't help. frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] Ceph: Monitors not running but cannot be destroyed or recreated
Dear all, I am trying to destroy "old" Ceph monitors but they can't be deleted and also cannot be recreated: I am currently configuring Ceph on our PVE cluster (3 nodes running PVE 6.1-3). There have been some "remainders" of a previous Ceph configuration which I had tried to configure while the nodes were not in a cluster configuration yet (and I had used the wrong network). However I had purged these configurations with `pveceph purge`. I have redone the basic Ceph configuration through the GUI on the first node and I have deleted the still existing managers through the GUI (to have a fresh start). A new monitor has been created on the first node automatically, but I am unable to delete the monitors on nodes 2 and 3. They show up as Status=stopped and Address=Unknown in the GUI and they cannot be started (no error message). In the syslog window I see (after rebooting node odcf-pve02): Jan 26 13:51:53 odcf-pve02 systemd[1]: Started Ceph cluster monitor daemon. Jan 26 13:51:55 odcf-pve02 ceph-mon[1372]: 2020-01-26 13:51:55.450 7faa98ab9280 -1 mon.odcf-pve02@0(electing) e1 failed to get devid for : fallback method has serial ''but no model On the other hand I see the same message on the first node, and there the monitor seems to work fine. Trying to destroy them results in the message, that there is no such monitor, and trying to create a new monitor on these nodes results in the message, that the monitor already exists I am stuck in this existence loop. Destroying or creating them also doesn't work on the commandline. Any idea on how to fix this? I'd rather not completely reinstall the nodes :-) Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] PVE Cluster: New authentication is required to access each node from GUI
The time it was. It was off by a few minutes between two of the servers but off by several hours on the third. I don't like ntpd anyway and I will probably replace it by chronyd. Thanks for the hint. Cheers, frank On 25/01/2020 18:32, Gianni Milo wrote: Things I would check or modify... - output of 'pvecm s' and 'pvecm n' commands. - syslog on each node for any clues. - ntp. - separate cluster (corosync) network from storage network (i.e In your case, use --link2, LAN). G. On Sat, 25 Jan 2020 at 15:44, Frank Thommen wrote: Dear all, I have installed a 3-node PVE cluster as instructed on https://pve.proxmox.com/pve-docs/chapter-pvecm.html (usung commandline). When I now connect via GUI to one node and select one of the other nodes, I get a "401" error message and then I am asked to authenticate to the other node. So to see all nodes from all other nodes via GUI I would have to authenticate nine times. I don't think that is as it should be ;-). I would assume that once I am logged in on the GUI of one of the cluster nodes, I can look at the other two nodes w/o additional authentication from this GUI. The situation is somehow similar to the one described on https://forum.proxmox.com/threads/3-node-cluster-permission-denied-invalid-pve-ticket-401.56038/, but the suggested "pvecm updatecerts" (run on each node) only helped for a short time. After a reboot of the nodes I am back to the potential nine authentications. My three nodes are connected through a full 10GE mesh (https://pve.proxmox.com/wiki/Full_Mesh_Network_for_Ceph_Server) using broadcast bonds. This mesh will finally also be used for Ceph. I configured this mesh to be the cluster network (--link0). As fallback (--link1) I used the regular LAN. Does anyone have an idea what could be wrong and how this could be fixed? Could the mesh with the broadcast bonds be the problem? If yes, should I use an other type of mesh? Unfortunately a full dedicated PVE-only network with a switch is not an option. I can either use a mesh or the regular LAN in the datacenter. The systems are running PVE 6.1-3. Any help or hint is appreciated. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] PVE Cluster: New authentication is required to access each node from GUI
Dear all, I have installed a 3-node PVE cluster as instructed on https://pve.proxmox.com/pve-docs/chapter-pvecm.html (usung commandline). When I now connect via GUI to one node and select one of the other nodes, I get a "401" error message and then I am asked to authenticate to the other node. So to see all nodes from all other nodes via GUI I would have to authenticate nine times. I don't think that is as it should be ;-). I would assume that once I am logged in on the GUI of one of the cluster nodes, I can look at the other two nodes w/o additional authentication from this GUI. The situation is somehow similar to the one described on https://forum.proxmox.com/threads/3-node-cluster-permission-denied-invalid-pve-ticket-401.56038/, but the suggested "pvecm updatecerts" (run on each node) only helped for a short time. After a reboot of the nodes I am back to the potential nine authentications. My three nodes are connected through a full 10GE mesh (https://pve.proxmox.com/wiki/Full_Mesh_Network_for_Ceph_Server) using broadcast bonds. This mesh will finally also be used for Ceph. I configured this mesh to be the cluster network (--link0). As fallback (--link1) I used the regular LAN. Does anyone have an idea what could be wrong and how this could be fixed? Could the mesh with the broadcast bonds be the problem? If yes, should I use an other type of mesh? Unfortunately a full dedicated PVE-only network with a switch is not an option. I can either use a mesh or the regular LAN in the datacenter. The systems are running PVE 6.1-3. Any help or hint is appreciated. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] How to get rid of default thin-provisioned default LV "data"?
That's what I have written in our internal documentation: -- We delete the current thin-provisioned LV, create a new "thick-provisioned" LV instead, format it, mount it and configure it as "dir" storage for PVE: First manually remove the lvmthin entry from /etc/pve/storage.cfg, then $ lvremove /dev/pve/data Do you really want to remove and DISCARD active logical volume pve/data? [y/n]: yes Logical volume "data" successfully removed $ lvcreate -n data -l 100%FREE pve Logical volume "data" created. $ mkfs -t ext4 /dev/pve/data mke2fs 1.44.5 (15-Dec-2018) Creating filesystem with 216501248 4k blocks and 54132736 inodes Filesystem UUID: 5d457f4c-40a5-4423-9c3a-03db1ae76869 Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000, 7962624, 11239424, 2048, 23887872, 71663616, 78675968, 10240, 214990848 Allocating group tables: done Writing inode tables: done Creating journal (262144 blocks): done Writing superblocks and filesystem accounting information: done $ mkdir /data $ echo "/dev/pve/data /data ext4 defaults 0 2" >> /etc/fstab && mount -a $ pvesm add dir data --path /data --content images,rootdir,vztmpl,iso,backup,snippets $ done... -- HTH, frank On 19.01.20 00:53, Saint Michael wrote: can somebody post the full chain of commands? That is the only reason why I am not using the software. On Sat, Jan 18, 2020 at 2:54 PM Frank Thommen wrote: Thanks Gianni, that worked just fine. I've now created a regular LVM LV instead, formatted and mounted it and configured it as "dir" storage. frank On 16.01.20 21:19, Gianni Milo wrote: This should be a case of removing (or comment out) its corresponding entry in /etc/pve/storage.cfg and then removing it by using the usual lvm commands. You can then "convert" it to thick lvm and re-add it in the config file... Gianni On Thu, 16 Jan 2020 at 18:26, Frank Thommen < f.thom...@dkfz-heidelberg.de> wrote: Dear all, when installing PVE on our servers with 1 TB boot disk, the installer creates regular /root and swap LVs (ca. 100 GB in total) and the rest of the disk (800 GB) is used for a thin-provisioned LV named "data". Personally I don't like thin-provisioning and would like to get rid of it and have a "regular" LV instead. However I couldn't find a way to achieve this through the gui (PVE 6.1-3). Did I miss some gui functionality or is it ok to remove the LV with regular lvm commands? Will I break something if I remove it? No VM/container has been installed so far. The hypervisors are freshly installed and they contain additional several internal disks which we plan to use for local storage/Ceph. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | f.thom...@dkfz-heidelberg.de ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | f.thom...@dkfz-heidelberg.de ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] How to get rid of default thin-provisioned default LV "data"?
Thanks Gianni, that worked just fine. I've now created a regular LVM LV instead, formatted and mounted it and configured it as "dir" storage. frank On 16.01.20 21:19, Gianni Milo wrote: This should be a case of removing (or comment out) its corresponding entry in /etc/pve/storage.cfg and then removing it by using the usual lvm commands. You can then "convert" it to thick lvm and re-add it in the config file... Gianni On Thu, 16 Jan 2020 at 18:26, Frank Thommen wrote: Dear all, when installing PVE on our servers with 1 TB boot disk, the installer creates regular /root and swap LVs (ca. 100 GB in total) and the rest of the disk (800 GB) is used for a thin-provisioned LV named "data". Personally I don't like thin-provisioning and would like to get rid of it and have a "regular" LV instead. However I couldn't find a way to achieve this through the gui (PVE 6.1-3). Did I miss some gui functionality or is it ok to remove the LV with regular lvm commands? Will I break something if I remove it? No VM/container has been installed so far. The hypervisors are freshly installed and they contain additional several internal disks which we plan to use for local storage/Ceph. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | f.thom...@dkfz-heidelberg.de ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] How to get rid of default thin-provisioned default LV "data"?
Dear all, when installing PVE on our servers with 1 TB boot disk, the installer creates regular /root and swap LVs (ca. 100 GB in total) and the rest of the disk (800 GB) is used for a thin-provisioned LV named "data". Personally I don't like thin-provisioning and would like to get rid of it and have a "regular" LV instead. However I couldn't find a way to achieve this through the gui (PVE 6.1-3). Did I miss some gui functionality or is it ok to remove the LV with regular lvm commands? Will I break something if I remove it? No VM/container has been installed so far. The hypervisors are freshly installed and they contain additional several internal disks which we plan to use for local storage/Ceph. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] LVM/pvesm udev database initialization warnings after import of KVM images
Dear all, after having (successfully) imported two KVM disk images from oVirt, LVM and pvesm complain about some udev initialization problem: root@pve01:~# pvesm status WARNING: Device /dev/dm-8 not initialized in udev database even after waiting 1000 microseconds. WARNING: Device /dev/dm-9 not initialized in udev database even after waiting 1000 microseconds. WARNING: Device /dev/dm-8 not initialized in udev database even after waiting 1000 microseconds. WARNING: Device /dev/dm-9 not initialized in udev database even after waiting 1000 microseconds. Name Type Status TotalUsed Available% local dir active9855922045640836 47868836 46.31% local-lvm lvmthin active 83224576071573135 7606726248.60% root@pve01:~# root@pve01:~# lvdisplay WARNING: Device /dev/dm-8 not initialized in udev database even after waiting 1000 microseconds. WARNING: Device /dev/dm-9 not initialized in udev database even after waiting 1000 microseconds. --- Logical volume --- LV Path/dev/pve/swap LV Nameswap VG Namepve [...] root@pve01:~# However this doesn't seem to influence the functionality of the VMs. Any idea what could be the problem and how to fix it? Thank you very much in advance Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Are two bridges to the same network possible?
On 12/17/19 10:35 PM, Chris Hofstaedtler | Deduktiva wrote: * Frank Thommen [191217 22:10]: I probably don't understand the question correctly: One has the IP traffic to the server/hypervisor, the other has IP traffic for the attached VMs and containers. Is there no need for a gateway if the bridge only provisions VMs and containers with network connectivity? Should I think of the bridge as a dumb connector on the link level (I should probably and urgently reread my networking materials :-) You can think of it as a dumb switch. As a special feature, Linux allows you to also setup host network connectivity on a bridge - but it doesn't have to do that. I would guess that on the bridge you want to use for your VMs, you don't need an IP adress (and also no gateway then). Generally speaking, having more than one default gateway per host is an advanced configuration and you really need to know what you're doing then. Having multiple interfaces with IP addresses is a common thing, but you'llneed to understand how your routing setup works. Thanks, that helps. I think I will additionally remove vmbr0 (using bond0), as this connection will not be used for any virtual machine or container, but only to access the hypervisor itself. frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Are two bridges to the same network possible?
On 17.12.19 21:42, Chris Hofstaedtler | Deduktiva wrote: * Frank Thommen [191217 20:29]: (how) can I connect two virtual bridges to the same network (using the same gateway)? Our currently-being-built-up PVE server has four Linux network bonds connecting to two three networks. One is being used for the PVE server itself (administration, web ui ecc), the other three for virtual machines which will reside in three different networks. One of these three networks is the same as the one where the PVE server is residing, but I'd still like to use separate NICs for the VMs. The server itself is attached to the default vmbr0 (using bond0). But as soon as I want to configure a second bridge vmbr1 (using bond1) to the same network, PROXMOX complains that the default gateway already exists. Is it technically possible/supported to have multiple bridges to the same network (with the same gateways)? Do you actually need IP traffic on the host on the second interface? If not, then don't configure IP addresses, gateways, etc... Chris I probably don't understand the question correctly: One has the IP traffic to the server/hypervisor, the other has IP traffic for the attached VMs and containers. Is there no need for a gateway if the bridge only provisions VMs and containers with network connectivity? Should I think of the bridge as a dumb connector on the link level (I should probably and urgently reread my networking materials :-) Cheers, frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] Are two bridges to the same network possible?
Hello, (how) can I connect two virtual bridges to the same network (using the same gateway)? Our currently-being-built-up PVE server has four Linux network bonds connecting to two three networks. One is being used for the PVE server itself (administration, web ui ecc), the other three for virtual machines which will reside in three different networks. One of these three networks is the same as the one where the PVE server is residing, but I'd still like to use separate NICs for the VMs. The server itself is attached to the default vmbr0 (using bond0). But as soon as I want to configure a second bridge vmbr1 (using bond1) to the same network, PROXMOX complains that the default gateway already exists. Is it technically possible/supported to have multiple bridges to the same network (with the same gateways)? Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] (Very) basic question regarding PVE Ceph integration
On 12/17/18 9:23 AM, Eneko Lacunza wrote: Hi, El 16/12/18 a las 17:16, Frank Thommen escribió: I understand that with the new PVE release PVE hosts (hypervisors) can be used as Ceph servers. But it's not clear to me if (or when) that makes sense. Do I really want to have Ceph MDS/OSD on the same hardware as my hypervisors? Doesn't that a) accumulate multiple POFs on the same hardware and b) occupy computing resources (CPU, RAM), that I'd rather use for my VMs and containers? Wouldn't I rather want to have a separate Ceph cluster? The integration of Ceph services in PVE started with Proxmox VE 3.0. With PVE 5.3 (current) we added CephFS services to the PVE. So you can run a hyper-converged Ceph with RBD/CephFS on the same servers as your VM/CT. a) can you please be more specific in what you see as multiple point of failures? not only I run the hypervisor which controls containers and virtual machines on the server, but also the fileservice which is used to store the VM and container images. I think you have less points of failure :-) because you'll have 3 points (nodes) of failure in an hyperconverged scenario and 6 in a separate virtualization/storage cluster scenario... it depends how you look at it. Right, but I look at it from the service side: one hardware failure -> one service affected vs. one hardware failure -> two service affected. b) depends on the workload of your nodes. Modern server hardware has enough power to be able to run multiple services. It all comes down to have enough resources for each domain (eg. Ceph, KVM, CT, host). I recommend to use a simple calculation for the start, just to get a direction. In principle: ==CPU== core='CPU with HT on' * reserve a core for each Ceph daemon (preferable on the same NUMA as the network; higher frequency is better) * one core for the network card (higher frequency = lower latency) * rest of the cores for OS (incl. monitoring, backup, ...), KVM/CT usage * don't overcommit ==Memory== * 1 GB per TB of used disk space on an OSD (more on recovery) Note this is not true anymore with Bluestore, because you have to add cache space into account (1GB for HDD and 3GB for SSD OSDs if I recall correctly.), and also currently OSD processes aren't that good with RAM use accounting... :) * enough memory for KVM/CT * free memory for OS, backup, monitoring, live migration * don't overcommit ==Disk== * one OSD daemon per disk, even disk sizes throughout the cluster * more disks, more hosts, better distribution ==Network== * at least 10 GbE for storage traffic (more the better), see our benchmark paper https://forum.proxmox.com/threads/proxmox-ve-ceph-benchmark-2018-02.41761/ 10Gbit helps a lot with latency; small clusters can work perfectly with 2x1Gbit if they aren't latency-sensitive (we have been running a handfull of those for some years now). I will keep the two points in mind. Thank you. frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] (Very) basic question regarding PVE Ceph integration
Hi Alwin, On 12/16/18 7:47 PM, Alwin Antreich wrote: On Sun, Dec 16, 2018 at 05:16:50PM +0100, Frank Thommen wrote: Hi Alwin, On 16/12/18 15:39, Alwin Antreich wrote: Hello Frank, On Sun, Dec 16, 2018 at 02:28:19PM +0100, Frank Thommen wrote: Hi, I understand that with the new PVE release PVE hosts (hypervisors) can be used as Ceph servers. But it's not clear to me if (or when) that makes sense. Do I really want to have Ceph MDS/OSD on the same hardware as my hypervisors? Doesn't that a) accumulate multiple POFs on the same hardware and b) occupy computing resources (CPU, RAM), that I'd rather use for my VMs and containers? Wouldn't I rather want to have a separate Ceph cluster? The integration of Ceph services in PVE started with Proxmox VE 3.0. With PVE 5.3 (current) we added CephFS services to the PVE. So you can run a hyper-converged Ceph with RBD/CephFS on the same servers as your VM/CT. a) can you please be more specific in what you see as multiple point of failures? not only I run the hypervisor which controls containers and virtual machines on the server, but also the fileservice which is used to store the VM and container images. Sorry, I am still not quite sure, what your question/concern is. Failure tolerance needs to be planned into the system design, irrespective of service distribution. Proxmox VE has a HA stack that restarts all services from a failed node (if configured) on a other node. https://pve.proxmox.com/pve-docs/chapter-ha-manager.html Ceph does selfhealing (if enough nodes are available) or still works in a degraded state. http://docs.ceph.com/docs/luminous/start/intro/ Yes, I am aware of PVE and Ceph failover/healing capabilities. But I always liked to separate basic and central services on the hardware level. This way if one server "explodes", only one service is affected. With PVE+Ceph on one node, such an outage would affect two basic services at once. I don't say they wouldn't continue to run productively, but they would run in degraded and non-failure-safe mode - assumed we had three such nodes in the cluster - until the broken node can be restored. But that's probably just my old-fashioned conservative approach. That's why I wanted to ask the list members for their assessment ;-) [...] Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] (Very) basic question regarding PVE Ceph integration
Hi Alwin, On 16/12/18 15:39, Alwin Antreich wrote: Hello Frank, On Sun, Dec 16, 2018 at 02:28:19PM +0100, Frank Thommen wrote: Hi, I understand that with the new PVE release PVE hosts (hypervisors) can be used as Ceph servers. But it's not clear to me if (or when) that makes sense. Do I really want to have Ceph MDS/OSD on the same hardware as my hypervisors? Doesn't that a) accumulate multiple POFs on the same hardware and b) occupy computing resources (CPU, RAM), that I'd rather use for my VMs and containers? Wouldn't I rather want to have a separate Ceph cluster? The integration of Ceph services in PVE started with Proxmox VE 3.0. With PVE 5.3 (current) we added CephFS services to the PVE. So you can run a hyper-converged Ceph with RBD/CephFS on the same servers as your VM/CT. a) can you please be more specific in what you see as multiple point of failures? not only I run the hypervisor which controls containers and virtual machines on the server, but also the fileservice which is used to store the VM and container images. b) depends on the workload of your nodes. Modern server hardware has enough power to be able to run multiple services. It all comes down to have enough resources for each domain (eg. Ceph, KVM, CT, host). I recommend to use a simple calculation for the start, just to get a direction. In principle: ==CPU== core='CPU with HT on' * reserve a core for each Ceph daemon (preferable on the same NUMA as the network; higher frequency is better) * one core for the network card (higher frequency = lower latency) * rest of the cores for OS (incl. monitoring, backup, ...), KVM/CT usage * don't overcommit ==Memory== * 1 GB per TB of used disk space on an OSD (more on recovery) * enough memory for KVM/CT * free memory for OS, backup, monitoring, live migration * don't overcommit ==Disk== * one OSD daemon per disk, even disk sizes throughout the cluster * more disks, more hosts, better distribution ==Network== * at least 10 GbE for storage traffic (more the better), see our benchmark paper https://forum.proxmox.com/threads/proxmox-ve-ceph-benchmark-2018-02.41761/ * separate networks, cluster, storage, client traffic, additional for separate the migration network from any other * use to physical networks for corosync (ring0 & ring1) This list doesn't cover every aspect (eg. how much failure is allowed), but I think it is a good start. With the above points for the sizing of your cluster, the question of a separation of a hyper-converged service might be a little easier. Thanks a lot. This sure helps in our planning. frank -- Cheers, Alwin ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] (Very) basic question regarding PVE Ceph integration
Hi, I understand that with the new PVE release PVE hosts (hypervisors) can be used as Ceph servers. But it's not clear to me if (or when) that makes sense. Do I really want to have Ceph MDS/OSD on the same hardware as my hypervisors? Doesn't that a) accumulate multiple POFs on the same hardware and b) occupy computing resources (CPU, RAM), that I'd rather use for my VMs and containers? Wouldn't I rather want to have a separate Ceph cluster? Or didn't I get the point of the Ceph integration? Cheers Frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Cluster network via directly connected interfaces?
Good point. Thanks a lot frank On 11/22/2018 07:51 PM, Uwe Sauter wrote: FYI: I had such a thing working. What you need to keep in mind is that you should configure both interfaces per host on the same (software) bridge and keep STP on… that way when you loose the link from node A to node B the traffic will be going through node C. ++ | | | Node A br0 | | / \ | | eth0 eth1 | +--/---\-+ / \ +/--+ +-\+ | eth1 | | eth0 | | / | | \ | | br0--eth0-eth1--br0 | | Node B | | Node C | +---+ +--+ Am 22.11.18 um 19:42 schrieb Frank Thommen: What I /really/ meant was "but the throughput would /not/ be higher when using a switch"... On 11/22/2018 07:37 PM, Frank Thommen wrote: But the throughput would be higher when using a switch, would it? It's still just 1Gbit frank On 11/22/2018 07:34 PM, Mark Schouten wrote: Other than limited throughput, I can’t think of a problem. But limited throughput might cause unforeseen situations. Mark Schouten Op 22 nov. 2018 om 19:30 heeft Frank Thommen het volgende geschreven: Please excuse, if this is too basic, but after reading https://pve.proxmox.com/wiki/Cluster_Manager I wondered, if the cluster/corosync network could be built by directly connected network interfaces. I.e not like this: +---+ | pve01 |--+ +---+ | | +---+ ++ | pve02 |-| network switch | +---+ ++ | +---+ | | pve03 |--+ +---+ but like this: +---+ | pve01 |---+ +---+ | | | +---+ | | pve02 | | +---+ | | | +---+ | | pve03 |---+ +---+ (all connections 1Gbit, there are currently not plans to extend over three nodes) I can't see any drawback in that solution. It would remove one layer of hardware dependency and potential spof (the switch). If we don't trust the interfaces, we might be able to configure a second network with the three remaining interfaces. Is such a "direct-connection" topology feasible? Recommended? Strictly not recommended? I am currently just planning and thinking and there is no cluster (or even a PROXMOX server) in place. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Cluster network via directly connected interfaces?
What I /really/ meant was "but the throughput would /not/ be higher when using a switch"... On 11/22/2018 07:37 PM, Frank Thommen wrote: But the throughput would be higher when using a switch, would it? It's still just 1Gbit frank On 11/22/2018 07:34 PM, Mark Schouten wrote: Other than limited throughput, I can’t think of a problem. But limited throughput might cause unforeseen situations. Mark Schouten Op 22 nov. 2018 om 19:30 heeft Frank Thommen het volgende geschreven: Please excuse, if this is too basic, but after reading https://pve.proxmox.com/wiki/Cluster_Manager I wondered, if the cluster/corosync network could be built by directly connected network interfaces. I.e not like this: +---+ | pve01 |--+ +---+ | | +---+ ++ | pve02 |-| network switch | +---+ ++ | +---+ | | pve03 |--+ +---+ but like this: +---+ | pve01 |---+ +---+ | | | +---+ | | pve02 | | +---+ | | | +---+ | | pve03 |---+ +---+ (all connections 1Gbit, there are currently not plans to extend over three nodes) I can't see any drawback in that solution. It would remove one layer of hardware dependency and potential spof (the switch). If we don't trust the interfaces, we might be able to configure a second network with the three remaining interfaces. Is such a "direct-connection" topology feasible? Recommended? Strictly not recommended? I am currently just planning and thinking and there is no cluster (or even a PROXMOX server) in place. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | frank.thom...@uni-heidelberg.de | MMK: +49-6221-54-3637 (Mo-Mi, Fr) | IPMB: +49-6221-54-5823 (Do) ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Cluster network via directly connected interfaces?
But the throughput would be higher when using a switch, would it? It's still just 1Gbit frank On 11/22/2018 07:34 PM, Mark Schouten wrote: Other than limited throughput, I can’t think of a problem. But limited throughput might cause unforeseen situations. Mark Schouten Op 22 nov. 2018 om 19:30 heeft Frank Thommen het volgende geschreven: Please excuse, if this is too basic, but after reading https://pve.proxmox.com/wiki/Cluster_Manager I wondered, if the cluster/corosync network could be built by directly connected network interfaces. I.e not like this: +---+ | pve01 |--+ +---+ | | +---+ ++ | pve02 |-| network switch | +---+ ++ | +---+ | | pve03 |--+ +---+ but like this: +---+ | pve01 |---+ +---+ | | | +---+ | | pve02 | | +---+ | | | +---+ | | pve03 |---+ +---+ (all connections 1Gbit, there are currently not plans to extend over three nodes) I can't see any drawback in that solution. It would remove one layer of hardware dependency and potential spof (the switch). If we don't trust the interfaces, we might be able to configure a second network with the three remaining interfaces. Is such a "direct-connection" topology feasible? Recommended? Strictly not recommended? I am currently just planning and thinking and there is no cluster (or even a PROXMOX server) in place. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | frank.thom...@uni-heidelberg.de | MMK: +49-6221-54-3637 (Mo-Mi, Fr) | IPMB: +49-6221-54-5823 (Do) ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] Cluster network via directly connected interfaces?
Please excuse, if this is too basic, but after reading https://pve.proxmox.com/wiki/Cluster_Manager I wondered, if the cluster/corosync network could be built by directly connected network interfaces. I.e not like this: +---+ | pve01 |--+ +---+ | | +---+ ++ | pve02 |-| network switch | +---+ ++ | +---+ | | pve03 |--+ +---+ but like this: +---+ | pve01 |---+ +---+ | | | +---+ | | pve02 | | +---+ | | | +---+ | | pve03 |---+ +---+ (all connections 1Gbit, there are currently not plans to extend over three nodes) I can't see any drawback in that solution. It would remove one layer of hardware dependency and potential spof (the switch). If we don't trust the interfaces, we might be able to configure a second network with the three remaining interfaces. Is such a "direct-connection" topology feasible? Recommended? Strictly not recommended? I am currently just planning and thinking and there is no cluster (or even a PROXMOX server) in place. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] HTTPS for download.proxmox.com
On 11/30/2017 03:11 PM, lemonni...@ulrar.net wrote: This is dumb. I agree that it wouldn't cost them anything to setup HTTPS, but I also agree that it is useless. The packages are signed and apt already checks the signature, HTTPS wouldn'd add anything at all. Not true: It gives you the certainty to be connected to the "real" proxmox page and not a fake page, e.g. by being redirected through a hacked nameserver or local resolver. And afaik, those using the community version don't have access to the enterprise repos. frank Unless you want to hide the fact that you are installing proxmox itself, but the connection to proxmox's repo itself gives that away. On Thu, Nov 30, 2017 at 03:01:53PM +0100, John Crisp wrote: On 30/11/17 14:34, Dietmar Maurer wrote: On 11/30/2017 02:21 PM, Dietmar Maurer wrote: I greatly respect the work you do on Proxmox but this specific response is under your habitual standards from a security standpoint. Exactly. That is why we provide the enterprise repository. IMHO the times where security and confidentiality (https) are limited to enterprise/paid services are long gone. As the OP noted, https comes at no cost and there is no reason not to have it configured. I'd even say, that https is mandatory for every site publishing more than just personal statements. Again, please use the enterprise repository if you want https. ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | frank.thom...@uni-heidelberg.de | MMK: +49-6221-54-3637 (Mo-Mi, Fr) | IPMB: +49-6221-54-5823 (Do) ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] HTTPS for download.proxmox.com
On 11/30/2017 02:21 PM, Dietmar Maurer wrote: I greatly respect the work you do on Proxmox but this specific response is under your habitual standards from a security standpoint. Exactly. That is why we provide the enterprise repository. IMHO the times where security and confidentiality (https) are limited to enterprise/paid services are long gone. As the OP noted, https comes at no cost and there is no reason not to have it configured. I'd even say, that https is mandatory for every site publishing more than just personal statements. Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] CT is locked (backup) - qm unlock doesn't help
On 01/12/2017 11:30 AM, Fabian Grünbichler wrote: On Thu, Jan 12, 2017 at 11:27:24AM +0100, Frank Thommen wrote: Hi, one of my LXC containers is locked: It cannot be started and cannot be backed up. The error message is "CT is locked (backup)". The only suggestion I found to solve this prolem is `qm unlock XXX`. However in my case, this results in the error message Configuration file 'nodes/pve02/qemu-server/XX.conf' does not exist yes, because "qm" is for qemu VMs. What else can I try to unlock this container? use the container tool "pct" -> "man pct" or "pct help" too easy :-). Everything works again. Thanks lot. frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Which service is used to mount datacenter NFS Share ?
On 01/04/2017 01:46 PM, Florent B wrote: On 01/04/2017 01:42 PM, Fábio Rabelo wrote: Hellows ... Look at storage view, then add . One of the options are nfs, you need to supply ip of the share, or, if it hass a dns name, you can use it . Fábio Rabelo 2017-01-04 10:38 GMT-02:00 Florent B: Hi everyone, I couldn't find the answer, I have a NFS share configured in datacenter storage. Which service (or systemd unit) is used to mount (and unmount) these shares ? Thank you. Florent ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user Hi Fábio, Excuse me if my message was not clear enough, but that was not my question. I already configured a NFS share. I just want to know, when nodes boot, which service (system service) is mounting those shares ? There is no service involved. You need the mounts to be configured in the mount table (/etc/fstab or automounter files) and the NFS client tools (mount.nfs & co.) need to be present. frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Help for new server
On 10/28/2016 09:21 AM, Eneko Lacunza wrote: Hi, El 27/10/16 a las 15:07, Frank Thommen escribió: I also need the battery for smart arrays? If you do not plan to use ZFS, which sort of nullifies the use of a RAID controller, then yes. If you care for your data a BBU is absolutely vital and will give you a noticeable performance boost. I understand that BBU improves performance, but in what way does it "care for your data - absolutely vital"? My understanding is that in fact it does add some risk to your data if RAID controller fails - bye bye write cache data! The BBU will store your unwritten cached data in case of a power failure and write it down after power has been restored. W/o BBU, you would loose that data completely. Of course, if you don't use cache, you also don't need the BBU ;-) Sure, but BBU is not vital for data care. If you want performance, you need BBU, but you can safely use write-through without BBU. BBU does not improve the consistency of your data. Your argumentation is not completely correct: If you want performanace, you need caching (not BBU). And if you use caching, you should really, really use BBU (your RAID controller might not even allow caching w/o BBU anyway) And yes ist does improve data consistency very much: In the case of a power failure. frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Help for new server
On 10/27/2016 10:05 AM, Eneko Lacunza wrote: Hi, Not trolling, just curious about a thing you said... El 26/10/16 a las 18:20, Michael Rasmussen escribió: I also need the battery for smart arrays? If you do not plan to use ZFS, which sort of nullifies the use of a RAID controller, then yes. If you care for your data a BBU is absolutely vital and will give you a noticeable performance boost. I understand that BBU improves performance, but in what way does it "care for your data - absolutely vital"? My understanding is that in fact it does add some risk to your data if RAID controller fails - bye bye write cache data! The BBU will store your unwritten cached data in case of a power failure and write it down after power has been restored. W/o BBU, you would loose that data completely. Of course, if you don't use cache, you also don't need the BBU ;-) And if the RAID controller is broken, you have a big problem anyway, no matter if you use caching or not ;-) frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Manager Skins/Themes
On 10/04/2016 10:47 AM, John Crisp wrote: On 03/10/16 21:12, Brian :: wrote: [...] The double vertical is just insane - I can't think of anywhere else that uses it. It just grates every time I look at it. There is no separation from the first column. We are all used to a vertical column left, and a horizontal row, in multiple programs (think say email programs, file managers, and just about anything else). With wide screens there is no need to try and cram everything to the left. If double columns were a good GUI choice you would see them everywhere, but you don't, which means GUI designers the world over consider them a bad design and do not use them. Anyway, I guess I am in a minority of one and will be ignored. You are not alone. I am with you, even though I wouldn't have formulated it so harshly :-). frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] PVE 4.3: Not enough space for side-by-side performance graphs
On 09/30/2016 07:47 AM, Fabian Grünbichler wrote: On Thu, Sep 29, 2016 at 07:01:14PM +0200, Frank Thommen wrote: Hi all, the upgrade to PVE 4.3 introduced some nice features in the GUI, e.g. the nice CPU, memory and swap usage bars. But unfortunately it also brought back the screen space waste issue solved back in June http://pve.proxmox.com/pipermail/pve-user/2016-May/010371.html and http://pve.proxmox.com/pipermail/pve-user/2016-June/010583.html). By moving the menu from a horizontal to a vertical position, there is now - on a standard 24" screen - not enough horizontal space left to show the graphs side-by-side w/o zooming out. not sure what a standard 24" screen is? all of this does not dpeend on screen size, but on resolution.. on a FHD (1920x1080) screen, two graphs should fit. yes, that's what I meant with 'standard 24" screen'. Anyway: On my 1920x1080 the two graphs don't fit. Browser (Firefox) is in maximum zoom but not fullscreen. Is there an (upgrade-safe) way to make the frames slightly smaller, so that they fit side-by-side w/o being forced to zoom out? you can make the leftmost tree smaller by grabbing the divider, maybe this helps? It does, but that's a manual intervention (mouse) like zooming (keyboard) which I'd like to avoid, since it's the question of just a few pixels. Cheers frank anyway, dietmar's suggesting about wasting less space by moving the legends seems like a good one :) ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Upgrade to Proxmox VE 4.3
On 09/29/2016 06:34 PM, Luis G. Coralle wrote: Hi all, I'have a single node Proxmox VE 4.2-2 To upgrade from 4.2 to 4.3 I run: apt-get update apt-get dist-upgrade After the process the version is 4.2-2 ( same version ) Do I need a enterprise subscription? Do you have this deb http://download.proxmox.com/debian jessie pve-no-subscription repository in your apt sources list? frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] [Feature] Float Div's on the summary pages
Great. Thanks for the hint. f. On 06/07/2016 01:01 PM, Felix Brucker wrote: im using 4.2-11/2c626aa1 currently, but it was also present in at least 4.2-10, upgrade your system and clear browser cache Best regards Felix From: frank.thom...@uni-heidelberg.de To: pve-user@pve.proxmox.com Date: Tue, 7 Jun 2016 12:56:09 +0200 Subject: Re: [PVE-User] [Feature] Float Div's on the summary pages "My" graphs (PVE 4.2-5/7cf09667) /always/ stack onto each other, no matter how wide the window ist, leaving a wide, unused, blank column on the right side (see https://ibin.co/2jpiRv8X5iXY.png). /I/'d like to see PVE use this space instead of wasting it (i.e. like in the mockup viewable on https://ibin.co/2jpiguIVbBuq.png) frank On 06/06/2016 12:51 PM, Felix Brucker wrote: Hi, love seeing it got integrated, thanks :) Best regards Felix From: felix.bruc...@live.de To: pve-user@pve.proxmox.com Subject: [Feature] Float Div's on the summary pages Date: Thu, 12 May 2016 00:18:56 +0200 Hi all, i don't know if this is the correct place to ask, but it would be great if the div's from the summary pages would float left, so that they take up the whole width of the screen but stack above each other if there is not enough free space. If it's not it would be great if someone could point me in the right direction, the forum seems to not have some sort of "Feature Request" section. Best regards Felix Brucker ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | frank.thom...@uni-heidelberg.de | TP3: +49-6221-42-3562 (Mo+Di) | IPMB: +49-6221-54-5823 (Mi-Do) ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | DKFZ Heidelberg / B080 | System Administrator | f.thom...@dkfz-heidelberg.de | MMK: currently n.a. (Mo+Mi, Fr) | IPMB: +49-6221-54-5823 (Do) -- Frank Thommen | HD-HuB / DKFZ Heidelberg | frank.thom...@uni-heidelberg.de | TP3: +49-6221-42-3562 (Mo+Di) | IPMB: +49-6221-54-5823 (Mi-Do) ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] [Feature] Float Div's on the summary pages
"My" graphs (PVE 4.2-5/7cf09667) /always/ stack onto each other, no matter how wide the window ist, leaving a wide, unused, blank column on the right side (see https://ibin.co/2jpiRv8X5iXY.png). /I/'d like to see PVE use this space instead of wasting it (i.e. like in the mockup viewable on https://ibin.co/2jpiguIVbBuq.png) frank On 06/06/2016 12:51 PM, Felix Brucker wrote: Hi, love seeing it got integrated, thanks :) Best regards Felix From: felix.bruc...@live.de To: pve-user@pve.proxmox.com Subject: [Feature] Float Div's on the summary pages Date: Thu, 12 May 2016 00:18:56 +0200 Hi all, i don't know if this is the correct place to ask, but it would be great if the div's from the summary pages would float left, so that they take up the whole width of the screen but stack above each other if there is not enough free space. If it's not it would be great if someone could point me in the right direction, the forum seems to not have some sort of "Feature Request" section. Best regards Felix Brucker ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | frank.thom...@uni-heidelberg.de | TP3: +49-6221-42-3562 (Mo+Di) | IPMB: +49-6221-54-5823 (Mi-Do) ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] NFSv4 support for NFS storage?
Hi, new fileservers in our department are configured to require NFSv4 and do not support NFSv3 any more. That worked fine for PVE 3.x by patching the "values.options" assignment in /usr/share/pve-manager/ext?/pvemanagerlib.js. When applying the same patch to PVE 4.x, the storage is mounted correctly and also shown in the summary, but the status is inactive and `pvesm list ` returns "mount error: mount.nfs: /mnt/pve/ is busy or already mounted". In the content tab, the error "mount error: mount.nfs: /mnt/pve/ is busy or already mounted (500)" is shown. Is there any other patch which has to be applied to make NFSv4 working with PVE 4.x and is there already a timeframe, when NFSv4 will be supported out of the box? We are currently running "pve-manager/4.1-33/de386c1a (running kernel: 4.2.6-1-pve)". Cheers Frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] User experience issues with Proxmox' new webUI ("Summary" pages)
On 05/03/2016 10:24 AM, Emmanuel Kasper wrote: On 04/29/2016 11:54 AM, Frank Thommen wrote: Hi, I have updated one of our PVE systems from 4.1-1 to 4.1-33 in one go and now I am seeing an overhauled webUI (specifically the summary pages). While I like the zoomable graphs, I have several issues with the new UI: a) font and line spacing have changed and are now using more (IMHO too much) space on the screen while not becoming better readable. Screenspace is valuable and I'd like it to be used as effectively as possible b) the additional space used in the "Status" box and the bigger size of the graphs have pushed the second graph ("Memory usage") out of my screen (fairly standard 24", 1920x1080). The old layout (I'm comparing with a still running 3.4-6) allowed me to see all the cpu usage and the top half of the memory usage w/o any scrolling. This was very efficient and in most cases just the information I was looking for. To get the same information with 4.1-33 I have to scroll down for each vm/container. c) graph buildup is now "animated" but also considerably slower. The old layout/graphs allowed me to very quickly browse through all vms/containers and check cpu and memory "by eye" as text boxes and graphs showed up almost instantaneously. With 4.1-33 I have to wait 1-2 secs for each vm/container to show the graphs. does zooming out with your browswer help to see more content here ? Theoretically, but then everything becomes so tiny that I find it hard to see anything. I'd rather reshuffle the graphs, so that the big unused (=wasted) whitespace to the right of the graphs is also used by some of the timelines...or I have to buy a rotatable screen and use it in portrait mode :-) please note that without client side rendering of the graphs, it would not be possible to zoom in and out the graphs themselves and client rendering takes a bit more of time of course I understand that from the technical side. But still My client is not /so/ slow... d) when by mistake one of the fields in the "Status" box is selected, the page always "jumps" back as soon as the page has been scrolled down... Yes I could see that issue the text fields here should actually not be selectable can you please submit a bug report in https://bugzilla.proxmox.com/ for that ? Done: https://bugzilla.proxmox.com/show_bug.cgi?id=976 frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] User experience issues with Proxmox' new webUI ("Summary" pages)
Hi, I have updated one of our PVE systems from 4.1-1 to 4.1-33 in one go and now I am seeing an overhauled webUI (specifically the summary pages). While I like the zoomable graphs, I have several issues with the new UI: a) font and line spacing have changed and are now using more (IMHO too much) space on the screen while not becoming better readable. Screenspace is valuable and I'd like it to be used as effectively as possible b) the additional space used in the "Status" box and the bigger size of the graphs have pushed the second graph ("Memory usage") out of my screen (fairly standard 24", 1920x1080). The old layout (I'm comparing with a still running 3.4-6) allowed me to see all the cpu usage and the top half of the memory usage w/o any scrolling. This was very efficient and in most cases just the information I was looking for. To get the same information with 4.1-33 I have to scroll down for each vm/container. c) graph buildup is now "animated" but also considerably slower. The old layout/graphs allowed me to very quickly browse through all vms/containers and check cpu and memory "by eye" as text boxes and graphs showed up almost instantaneously. With 4.1-33 I have to wait 1-2 secs for each vm/container to show the graphs. d) when by mistake one of the fields in the "Status" box is selected, the page always "jumps" back as soon as the page has been scrolled down... These experiences have been made with openSuSE 13.2 and Firefox 45.0. Is there is a way to customize fonts/linespacing, graph sizes and graph "animations" in the webUI, as to restore its previous great speed and efficiency? On the other hand I think that the following two features could improve the overview's efficiency and usefulness a lot: a) a possibility to zoom into all graphs simultaneously (like Ctrl-zoom or similar) b) graphs which show average and max in a single graph Cheers frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] Debian 8 Template w/o systemd?
On 04/20/2016 06:14 PM, Dietmar Maurer wrote: I'm a little bit puzzled, that the PVE-Template for the Debian 8 container comes w/o systemd. Instead it looks like an SysV init system. Is there a specific reason for this? yes, systemd inside debian 8 is very old, and many users had problems in the past. Not sure if that is still the case. On a completely updated Debian 8 container on PVE 3.4 (started from Debian 7 and dist-upgraded to 8), systemd is version 215. According to https://github.com/systemd/systemd/releases?after=v220 this version is of July 3, 2014. I assume this /is/ considered old :-). On the other hand I haven't experienced any problems so far. On https://wiki.debian.org/systemd it says "Please make sure that you are using Debian jessie or newer to get a recent version of systemd.". Hmm, "recent"...? ;-) But not including systemd makes the task of streamlining an IT environment harder. If you try to keep bare metal and virtual hosts on the same distro and release (Debian 8), one can still not use the same mechanisms and settings everywhere (systemd on metal, SysV init in containers). Switching to systemd once a system has been installed as SysV init system hasn't worked for me (yet). frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] No 32bit container templates for PVE 4.x?
On 03/03/2016 11:14 PM, Ilya Konyakhin wrote: On Thu, Mar 3, 2016 at 9:31 PM, Frank Thommen <frank.thom...@uni-heidelberg.de <mailto:frank.thom...@uni-heidelberg.de>> wrote: What do you mean by "the previous generated 32 bit templates"? Should additional templates be downloadable somewhere? I couldn't find them on linuxcontainers.org <http://linuxcontainers.org> Frank, I suggest that Thomas said about openvz templates. You can find them here: http://wiki.openvz.org/Download/template/precreated Meaning that I can use openVZ templates for LXC containers? That's great. I had assumed, that the templates contain configurations or information specific to the selected container mechanism. Thanks for the clarification frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
Re: [PVE-User] No 32bit container templates for PVE 4.x?
Hi, thanks for the quick answer. On 03/03/2016 04:11 PM, Thomas Lamprecht wrote: Hi, no no reason at all, LXC and so we is able to run 32bit Containers fine. You can use the one provided by lxc (at least those which distro we support (debian, ubuntu, arch, centos, alpine, suse are the ones if I did not miss anything) and you normally also should be able to run the previous generated 32 bit templates just fine. What do you mean by "the previous generated 32 bit templates"? Should additional templates be downloadable somewhere? I couldn't find them on linuxcontainers.org Our dab and aab (debian and archlinux appliance builder) can generate 32 bit templates/appliances. So the reason that there are no 32 bit ones at our template host is unknown to me but probably simply nobody thought of this yet, as the overhead is often quite small and 64 bit has its advantages. OK. I might just switch to 64bit containers, then. Building (at least a debian, arch one) and upload it should be no problem, but no promise how fast :) Debian would the one I'm currently specifically looking for :-) Cheers Frank cheers, Thomas On 03.03.2016 14:09, Frank Thommen wrote: Hello, is there a reason, why there are no 32bit container templates provided for PVE 4.x? With PVE 3.x I used the 32bit openvz templates, as I'm mostly running small services and I want to keep the memory and disk footprints as small as possible. However for PVE 4.x all offered templates seem to be amd64. frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user -- Frank Thommen | HD-HuB / DKFZ Heidelberg | frank.thom...@uni-heidelberg.de | TP3: +49-6221-42-3562 (Mo+Di) | IPMB: +49-6221-54-5823 (Mi-Fr) ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
[PVE-User] No 32bit container templates for PVE 4.x?
Hello, is there a reason, why there are no 32bit container templates provided for PVE 4.x? With PVE 3.x I used the 32bit openvz templates, as I'm mostly running small services and I want to keep the memory and disk footprints as small as possible. However for PVE 4.x all offered templates seem to be amd64. frank ___ pve-user mailing list pve-user@pve.proxmox.com http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user