Archive for the ‘EN’ Category
Something that doesn’t show up anywhere in combination with a Pulseway monitoring sytem, sometimes the RDP session can be slow as snail OS.
What happened, a new server was set up to monitor client server and with some, trying to log into the server through the Pulseway Remote Desktop Client either was very slow, or it just hung at waiting for a screen update forever.
A colleague was investigating the issue together with support from Pulseway, who suspected the cause was the reverse proxy. The server had been moved directly behind the local router and the issue still persisted.
While having a look at it myself I discovered if I restricted RDP features on server side the session improved a little. Looking further I noticed the Pulseway client connects to the server console, not a standard Windows RDP session.
Something that I haven’t had to do in years fixed this issue, namely turning off graphic hardware acceleration. (right click desktop, click screen resolution, the link to advanced settings, tab troubleshooting and slide the slider all the way to the left.
The sessions were suddenly as fast as a normal RDP session.
This thing hasn’t been documented anywhere, so hopefully it’ll help if anyone who has the same slow RDP client sessions on other remote monitoring or shared desktop software like Teamviewer and ISL.
To work around the restriction of Adobe Lightroom refusing to write catalogs on a network share (for Windows PC);
- Connect the share as a normal network share on the Windows PC; for example drive letter Z:
- Create the desired directory on the share; for example Z:\Lightroom
- in a command promtp run subst x: z:\lightroom (or any free drive letter)
- Now you can create a new catalog on X:
The command subst x: z:\lightroom can be made as a shortcut, or even set up as a logon script in the start menu for all or single users.
Finally getting around to reinstall my fileserver, I wanted to make it flexible with data integrity.
ZFS wasn’t flexible enough for my tastes, and LVM has no integrity check, so I had to create something with btrfs.
Early tests with using btrfs on LVM for root and /boot made the virtual machine unbootable, but using btrfs directly was no problem.
This meant I still had to use separate partitions for /boot, root, and swap (on mdraid10), and then use the rest for LVM.
It seems the installation of Stretch has trouble when adding raid or lvm afterwards, so I partitioned the installation this way:
vda1, 512M, bootable, btrfs, /boot
vda2, 4096M, btrfs, /
vda3, 512M, device for raid (to be made swap after creating raid10 after installation)
vg0: vda4, vdb4, vdc4
tmp_0, 512M, btrfs, /tmp
home_0, 1024M, btrfs, /home
var_0, 1024M, btrfs, /var
After installation I created the same partitions for vdb and vdc, then created the mdraid10 for swap:
mdadm –create –run /dev/md99 –level=10 –layout=n2 raid-devices=3 /dev/vda3 /dev/vdb3 /dev/vdc3
With that done I exchanged all /dev/mapper devices for the UUID’s of the btrfs devices in /etc/fstab and did a reboot as test.
With everything running I created logical volumes tmp_1, tmp_2, home_1, home_2, etc. by specifying the different physical devices to be put on:
lvcreate -L 512M -n tmp_1 vg0 /dev/vdb4
lvcreate -L 512M -n tmp_2 vg0 /dev/vdc4
Adding the LV’s to btrfs:
btrfs device add -f /dev/mapper/vg0-tmp_1 /dev/mapper/vg0-tmp_2 /tmp
btrfs device add -f /dev/mapper/vg0-home_1 /dev/mapper/vg0-home_2 /home
btrfs balance start -dconvert=raid1 -mconvert=raid1 /tmp
btrfs balance start -dconvert=raid1 -mconvert=raid1 /home
For unimportant data I made a logical volume striped across the three disks:
lvcreate -L xxG -n roraid -i 3 vg0
mkfs.ext4 /dev/mapper/vg0-noraid (could also be btrfs but the data integrity isn’t important here.
..and added its UUID to fstab with mountpoint /data/noraid
For the data protected with checksum integrity:
create three LV’s raid1_0, 1, 2 on the three PV’s like /tmp, /home, etc. above and make another btrfs raid of these.
Last, don’t forget to grub-install /dev/vdb and /dev/vdc.
The /data/raid1 mount point caused an error after first reboot, but later no more. Maybe a glitch in the matrix..
With this setup I can expand /tmp, /home, /var, /data/noraid and /data/raid1 as needed since I don’t know if I get more data for raid or not.
With two fileservers I can keep a backup of the raid1 data of one server on the noraid volume of the other. The btrfs volumes only protect from bit rot after all.
And if it really fills up I can start exchanging the disks for bigger ones.
Social jetlag, the difference between the socially expected sleeping pattern and your true natural pattern is plain bad for your health. In my case, it feels destructive to mine with constant tiredness, bad short term memory and depressive tendencies.
Although they call it a disorder, I believe it’s a natural state since not every living creature is identical to others, and there is no proven definition of a human’s sleeping pattern (sleeping in two parts seemed common in medieval times).
So, after 17 years of working in IT I’m going to choose my health and happiness over social norms before it kills me.
I’m quitting and I’m not going to start my day before noon (emergencies excepted of course) anymore.
Back to creative and productive evenings for me.
After replacing the failing home server with a decent Synology NAS I got my hands on a HP Storageworks Data Vault X510.
Turns out it’s not too difficult.
-place the harddisk to install it on in a standard PC and use that to install the i386 version of Debian (in this case it was Jessie)
-add a fixed IP address for eth0 in /etc/networking/interfaces (the Realtek r8168 (seen as r8169) couldn’t get a DHCP address for some reason)
-place the disk in the X510 and boot
-login with SSH (you did install it, didn’t you?)
-have fun with it
We can set up a cluster with a virtual layer for operating systems and live migration, why not the same for programs?
Add a virtual layer between OS and programs and it would be possible to live migrate a running program from one cluster node to another, probably with less time and overhead than moving an entire OS, right?
The ones that mean something to me/Degenen die iets betekenen voor mij;
Nothing to fear but fear itself/Niks vrezen behalve de angst zelf (no real dutch translation)
Talk is easy, doing is hard/Spreken is zilver, doen is goud (vrije aanpassing op het bekende gezegde :-) )
Someone thought it was a good idea to use an HP Data Vault X510 as a backup device for a Windows 2003 SBS server. Well, not really the thing I’d choose but as it was already there I was going to use it.
First problem was not knowing the Administrator password, and of course no documentation on anything that thing does. :-/
Second problem was the missing recovery disc. No disc, no restoring the server and assigning a new password. Time for some oldfashioned hacking.
First off used the Hirens boot cd to blank the administrator password but that wasn’t enough to gain access as Windows doesn’t allow remote access without a password set. I pulled the first disk, hooked it up to another Windows machine and used regedit to load the SOFTWARE hive from (sysdrive):\windows\system32\config. There I added the DWORD value IsRemoteAccessUser to (hivename)\Microsoft\Windows Home Server\User Manager\Users\Administrator and set it to 1. This way I would be able to use Remote Desktop to connect to the Data Vault.
To actually be able to login I also needed to load the SYSTEM hive and change LimitBlankPasswordUse to 0 in (hivename)\ControlSet001\Control\Lsa and ControlSet003 (or 002).
After plugging the disk back in and booting up I was able to log into the server and set things right.
Even though I haven’ t tried any beta version I decided to go ahead and upgrade one of my cluster nodes to version 2.0 of Proxmox virtual environment.
The tricky part would most likely be the customized PVE installation I build; two disks running partially software RAID 1 with the system split in several partitions. That might break the upgrade script provided by Proxmox.
Actually, thing went pretty well so far. The biggest error was that I had previously left open a couple of remove assignments through aptitude which removed the pve kernel, so the script halted on that when it couldn’t find it.
Nothing a reversal to lenny in the sources.list and the install of the missing kernel couldn’t fix.
One thing that didn’t match up was not being able to start cman which complained about missing the nodename in cluster.conf. It was present in /etc/pve/cluster.conf but unfortunately the init script pointed to /etc/cluster/cluster.conf.
Adding the name to the latter conf was enough to start cman and enable me to log into the web interface. That one kept looping and timed out while the cluster deamon wasn’t running.
One thing te remember is that the web interface was listening to the default port again so I had to modify the Listen entry in pve.conf under the /etc/apache/sites directories.
Next is copying the backups from the old master node to the new and import them to see if they’ll run. When they do, the old master will be upgraded after confirming that the new master works as planned.
Just started the upgrade of the second node, so far I only had to do an apt-get install console-tools after an error mentioning that package during the apt-get upgrade round, and an apt-get install vzctl during the second attempt.
So, what to do when going from a job in truck assembly which meant 9 hours of body building a day to a job in IT sitting behind a keyboard which means slouching 10+ hours a day?
Eventually I noticed less vigor in taking the stairs so I looked around at various options. What I first chose was Tai Chi. It only took about two weeks of just practicing the first few forms to make me skip steps going up the stairs again.
Later on I went through a load of books on physical training and ended up with Pavel Tsatsouline’s books on kettlebell and body weight training (calisthenics). And to complement it there’s the book Convict Conditioning by Paul Wade.
What it comes down to is, with these resources and a little bit of time each day for the different excercises I manage to gain strength bit by bit again. Although it won’t be quick, there’s definitely an improvement I can recommend to anyone.