From zanchey at ucc.gu.uwa.edu.au Fri Feb 13 15:01:19 2015 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Fri, 13 Feb 2015 15:01:19 +0800 Subject: [tech] Meersau upgrade & RDP server Message-ID: <9fb28f1af13e300286b7c7fc38fad466@secure.ucc.asn.au> I upgraded Meersau to Jessie (pre-release Debian 8.0) to test some things out. I had to remove GNOME 3 in the process, due to a bug with the acceleration detection[1] that stopped remote access from working. Meersau is our VM that used to host the thinterms using LTSP (although I'm pretty sure they haven't booted properly for ages). I installed xrdp, a Microsoft Remote Desktop (RDP) server on it which works reasonably well. If you don't need a specific Windows tool, consider using Meersau for your remote desktop needs instead of Maaxen, which only has two concurrent users at once. You do need to SSH tunnel to access Meersau at the moment. David Adam UCC Wheel Member zanchey at ucc.gu.uwa.edu.au [1]: https://bugs.launchpad.net/ubuntu/+source/gnome-session/+bug/1251281 From oxinabox at ucc.asn.au Tue Feb 17 19:55:00 2015 From: oxinabox at ucc.asn.au (Frames) Date: Tue, 17 Feb 2015 19:55:00 +0800 Subject: [tech] Power Failure Message-ID: <54E32C14.6060206@ucc.asn.au> We had a unexpected power-failure on Red 1. It is over now. Molmol seemed to crash, seems to be back now. Frames From zanchey at ucc.gu.uwa.edu.au Fri Feb 20 14:45:16 2015 From: zanchey at ucc.gu.uwa.edu.au (David Adam) Date: Fri, 20 Feb 2015 14:45:16 +0800 (AWST) Subject: [tech] LDAP size limits exceeded In-Reply-To: References: Message-ID: Bumped up again... need a proper fix. On Mon, 28 Jul 2014, David Adam wrote: > Tonight, [ASH] was trying to add a new user. The membername.ucc.asn.au > alias wasn't being created correctly, and it turned out this was because > the "list all users" operation in Zonemake (analagous to `getent passwd`) > was hitting the maximum number of answers to a query in the LDAP server > (1000). > > I've bumped up the maxium size to 1500 (olcSizeLimit: 1500), but perhaps > we could consider: > > a) retiring some old usernames. `getent passwd | grep locked2004` might be > a good place to start; 10 years to reactivate should be enough. > b) enabling paged queries (setting `pagesize 500` or so in nslcd.conf on > all our clients, and setting `olcSizeLimit: 1500 size.prtotal=unlimited` > or similar > c) removing size limits on the server altogether; I don't think we're ever > going to have enough entries to seriously cause performance degradation. > > David Adam > UCC Wheel Member > zanchey at ucc.gu.uwa.edu.au > _______________________________________________ > List Archives: http://lists.ucc.gu.uwa.edu.au/pipermail/tech > > Unsubscribe here: http://lists.ucc.gu.uwa.edu.au/mailman/options/tech/zanchey%40ucc.gu.uwa.edu.au > > Cheers, David Adam zanchey at ucc.gu.uwa.edu.au Ask Me About Our SLA! From bob at ucc.gu.uwa.edu.au Sun Feb 22 20:59:15 2015 From: bob at ucc.gu.uwa.edu.au (Andrew Adamson) Date: Sun, 22 Feb 2015 20:59:15 +0800 (AWST) Subject: [tech] Medico outage update Message-ID: Hi all, Just a quick update on what happened to medico on Friday night. Proxmox needed to be upgraded from 2 (squeeze) to 3 (wheezy) because the web interface was becoming unusable on version 2. I also want to set up the VM cluster with maltair with all of them on the same version. Semester kinda snuck up on me and I wanted to get this done before it started, so sorry for the (very) short notice. Unfortunately during the upgrade, the new version of grub that was to be installed was too big to fit in the allocated space on the disk. I was at home at this point so tried to fix it remotely with little success. [TPG] managed to get it booting using grub on a usb stick on Saturday morning, and [DAA] is working on a permanent fix for it now. Aside from the bootloader, the rest of the upgrade went well, although no machines started at boot because my partial cluster configuration meant there was no quorum between only two hosts - they can be manually started though. I'll be working on finishing the cluster setup in the next week or so. Andrew Adamson bob at ucc.asn.au |"If you can't beat them, join them, and then beat them." | | ---Peter's Laws | From mjpomery at ucc.asn.au Sun Feb 22 22:21:23 2015 From: mjpomery at ucc.asn.au (Mitchell Pomery) Date: Sun, 22 Feb 2015 22:21:23 +0800 (AWST) Subject: [tech] Medico outage update In-Reply-To: References: Message-ID: One of the "features" Proxmox seems to have implemented is a notification on login about subscriptions. "You do not have a valid subscription for this server. Please visit www.proxmox.com to get a list of available options." Ignore it. All it is selling is support subscriptions, not licensing for the software itself. I haven't found a way to disable that message from popping up as of yet, but when I do, it will be gone. See the following link for more information about it. http://forum.proxmox.com/threads/15682-ProxmoxVE-will-change-LICENCE As for the fact proxmox still gives us cert errors on the web interface, I will look at fixing that tomorrow some time. Mitch On Sun, 22 Feb 2015, Andrew Adamson wrote: > Hi all, > > Just a quick update on what happened to medico on Friday night. Proxmox > needed to be upgraded from 2 (squeeze) to 3 (wheezy) because the web > interface was becoming unusable on version 2. I also want to set up the VM > cluster with maltair with all of them on the same version. Semester kinda > snuck up on me and I wanted to get this done before it started, so sorry > for the (very) short notice. > > Unfortunately during the upgrade, the new version of grub that was to be > installed was too big to fit in the allocated space on the disk. I was at > home at this point so tried to fix it remotely with little success. [TPG] > managed to get it booting using grub on a usb stick on Saturday morning, > and [DAA] is working on a permanent fix for it now. > > Aside from the bootloader, the rest of the upgrade went well, although no > machines started at boot because my partial cluster configuration meant > there was no quorum between only two hosts - they can be manually started > though. I'll be working on finishing the cluster setup in the next week or > so. > > Andrew Adamson > bob at ucc.asn.au > > |"If you can't beat them, join them, and then beat them." | > | ---Peter's Laws | > _______________________________________________ > List Archives: http://lists.ucc.gu.uwa.edu.au/pipermail/tech > > Unsubscribe here: http://lists.ucc.gu.uwa.edu.au/mailman/options/tech/bobgeorge33%40ucc.asn.au > From oxinabox at ucc.asn.au Mon Feb 23 09:31:49 2015 From: oxinabox at ucc.asn.au (Frames) Date: Mon, 23 Feb 2015 09:31:49 +0800 Subject: [tech] Medico outage update In-Reply-To: References: Message-ID: <54EA8305.8050309@ucc.asn.au> Isn't this the second time this has happened to you bob? :-P [*OX] "Don't blame, learn" On 22/02/2015 8:59 PM, Andrew Adamson wrote: > Hi all, > > Just a quick update on what happened to medico on Friday night. Proxmox > needed to be upgraded from 2 (squeeze) to 3 (wheezy) because the web > interface was becoming unusable on version 2. I also want to set up the VM > cluster with maltair with all of them on the same version. Semester kinda > snuck up on me and I wanted to get this done before it started, so sorry > for the (very) short notice. > > Unfortunately during the upgrade, the new version of grub that was to be > installed was too big to fit in the allocated space on the disk. I was at > home at this point so tried to fix it remotely with little success. [TPG] > managed to get it booting using grub on a usb stick on Saturday morning, > and [DAA] is working on a permanent fix for it now. > > Aside from the bootloader, the rest of the upgrade went well, although no > machines started at boot because my partial cluster configuration meant > there was no quorum between only two hosts - they can be manually started > though. I'll be working on finishing the cluster setup in the next week or > so. > > Andrew Adamson > bob at ucc.asn.au > > |"If you can't beat them, join them, and then beat them." | > | ---Peter's Laws | > _______________________________________________ > List Archives: http://lists.ucc.gu.uwa.edu.au/pipermail/tech > > Unsubscribe here: http://lists.ucc.gu.uwa.edu.au/mailman/options/tech/oxinabox%40ucc.asn.au From sam.moore at ucc.asn.au Mon Feb 23 17:43:40 2015 From: sam.moore at ucc.asn.au (Sam Moore) Date: Mon, 23 Feb 2015 17:43:40 +0800 Subject: [tech] Medico outage update In-Reply-To: <54EA8305.8050309@ucc.asn.au> References: <54EA8305.8050309@ucc.asn.au> Message-ID: <54EAF64C.7080709@ucc.asn.au> Perhaps we need to plot the size of the grub image against the version number to determine the rate at which to increase allocated space. [SZM] On 23/02/15 09:31, Frames wrote: > Isn't this the second time this has happened to you bob? > :-P > > [*OX] > "Don't blame, learn" > > On 22/02/2015 8:59 PM, Andrew Adamson wrote: >> Hi all, >> >> Just a quick update on what happened to medico on Friday night. Proxmox >> needed to be upgraded from 2 (squeeze) to 3 (wheezy) because the web >> interface was becoming unusable on version 2. I also want to set up the VM >> cluster with maltair with all of them on the same version. Semester kinda >> snuck up on me and I wanted to get this done before it started, so sorry >> for the (very) short notice. >> >> Unfortunately during the upgrade, the new version of grub that was to be >> installed was too big to fit in the allocated space on the disk. I was at >> home at this point so tried to fix it remotely with little success. [TPG] >> managed to get it booting using grub on a usb stick on Saturday morning, >> and [DAA] is working on a permanent fix for it now. >> >> Aside from the bootloader, the rest of the upgrade went well, although no >> machines started at boot because my partial cluster configuration meant >> there was no quorum between only two hosts - they can be manually started >> though. I'll be working on finishing the cluster setup in the next week or >> so. >> >> Andrew Adamson >> bob at ucc.asn.au >> >> |"If you can't beat them, join them, and then beat them." | >> | ---Peter's Laws | >> _______________________________________________ >> List Archives: http://lists.ucc.gu.uwa.edu.au/pipermail/tech >> >> Unsubscribe here: http://lists.ucc.gu.uwa.edu.au/mailman/options/tech/oxinabox%40ucc.asn.au > > _______________________________________________ > List Archives: http://lists.ucc.gu.uwa.edu.au/pipermail/tech > > Unsubscribe here: http://lists.ucc.gu.uwa.edu.au/mailman/options/tech/matches%40ucc.gu.uwa.edu.au >