There was a fail event reported on rosemary from one side of a pair of 60GB SSDs, which hold Rosemary_Data. Typical of my installations this mirror set holds the stuff the system needs beyond the os install: /home, databases, certificates, repositories, mail, samba, local bin, etc. Its a mirror set with an encrypted container, containing a btrfs filesystem. The older versions of these setups contain separate btrfs subvolumes for the different directories, newer ones have only one subvolume for that, and another for snaps. This is an old one.
Rosemary doesn’t have an extensive set of services – really only the /home and the databases. No real need for much of anything else. The local bin comes out of the repo anyway, there is no mail, no repository, no certs. However, without that volume the system won’t come up in a usable way. So lesson one learned here was when you get a fail event, attend to it. I let it go for a few days, because I knew I was going to have to pull the case out of the rack mount to get at the SSDs.
Continue reading Rosemary Recovery 2020
GDM is the display manager I’m using under Arch. I think it is the default DM in Arch, but I don’t really remember if I’ve changed it. In any case, the issue that has arisen is that when GDM is started by systemd after a reboot, it is launched with its own pulseaudio daemon. Then when I log in, as dee, I get a second pulseaudio daemon for dee (which is actually the desired one).
Most of the time this doesn’t matter, I guess. But I’m interested in enabling network sources in pulseaudio so I can (perhaps) have the boxes in the basement send their sound upstairs to oregano. Right now avahi is broadcasting two different network sound services on oregano, one from dee and one from gdm.
I want to stop pulseaudio from launching under gdm.
I thought this would be a simple matter of turning off autospawn in the client conf. The client conf is located in ~/.config/pulse (I think it used to be in ~/.pulse and was moved to be more “correct”). And ~ for gdm is /var/lib/gdm, so I tried client.conf in both .config/pulse/client.conf and .pulse/client.conf, but neither worked.
Poked around a little more and discovered that /usr/lib/systemd/user has a pulseaudio.service and pulseaudio.socket so these are actually being launched there by systemd. After a little reading I found that one could mask them by creating a local user override in ~/.config/systemd (for gdm this is /var/lib/gdm/.config/systemd), so I put in a /var/lib/gdm/.config/systemd/user/pulseaudio.socket and symlinked it to /dev/null.
And when I rebooted, sure enough I got not pulseaudio daemon for gdm.
It has become a constant annoyance that every time I build a new Raspberry Pi, the thing which is the most difficult is doing updates. I have scripts written to set everything up, but the first thing the script does is attempt to update software. Before I even reach that point, the first thing the Pi itself does is ask to update software. And it always fails.
I think this may be because of where I live. I think that the Raspbian (now Raspberry Pi OS apparently) mirrors must have one locally that is unreliable, and it gives me that one.
Two things I need to do. One is that it tries to connect using ipv6, and fails. I have to tell it to use ipv4. So one thing I end up having to do is tell apt not to use ipv6:
echo 'Acquire::ForceIPv4 "true";' > /etc/apt/apt.conf.d/99-useipv4
The second thing is to change the location of the mirror. One can look up the mirrors at https://www.raspbian.org/RaspbianMirrors, and pick one nearby, and replace the entry in /etc/apt/sources.list.
I first tried to install Mac OS onto a VM back in 2010 or 2011 I think. I’ve come back to the task from time to time and have never been successful, but truthfully until recently all the virtual machines I’ve created on Cinnamon had been too slow to be useful anyway. Now that I have Cinnamon doing virtual machines well, I came back to the Mac – at the same time I was doing other VMs, and I didn’t keep careful track of what I did to get it up. This document is meant to record what I remember of what I did the first time, and then to do it again and keep better records.
Continue reading Installing Mac OS on KVM
When I tried to upgrade Cinnamon to Focal, I began to experience a lot of odd problems with VNC and GDM and the whole collection of machinery associated with getting a graphical environment particularly in a remote window. Cinnamon is physically downstairs in a rack in the basement, and my usual way of working is 80-90% ssh/command line and occasionally vinagre in an adjacent monitor with screens for a dozen or so places that I occasionally need to see graphically, including Cinnamon, but Vinagre usually stays pointed at Rosemary (also in the basement).
Continue reading Packagekit adjustments in ubuntu
Some people for whom I provide some kinds of support with gateway pis, use Macs. For the pc folk – at least for those on Windows 10, I’ve been seting up to do the filehistory thing, and putting the filehistory onto the /backup drive on the gateway pi. Then it gets sent here overnight. I wanted to do the same for the folks who have Macs, of which there are several.
Continue reading Timemachine on Gateway pi
After updating Arch today I found I could no longer mount usb storage devices – usb sticks. The log shows the device coming up, but then messages from MTP saying:
mtp-probe: bus: 2, device: 5 was not an MTP device
I have seen this before. It is irritating, I am doing something else, my head full of whatever it is, and suddenly usb sticks won’t mount.
Continue reading libmtp udev rules
Shot off my foot.
I was working on some scripts which I will use on some other people’s macs (my wife has a new macbook air, and there is another friend I need to do this for also). These are bash scripts to be run from the terminal, and are supposed to set them up so I can help them from afar. One of the things the script does is create a user for me to use. I guess I was kinda “practicing” the transfer of the script via email, and I thoughtlessly ran the script on my macbook. Oops.
It tried to create a user dee with command line tools, and there was already a user dee. This was in El Capitan (the latest my ancient macbook will run), and there seemed to be no safeguards against idiocy. It seemed to foul things up. I rebooted, user dee was hidden — it was partly the new one, partly the old one, it wanted to “repair” the keychain, but couldn’t. There were a bunch of question marks in the dock. It couldn’t run anything. It was a mess. I was able to get up under a different username, and I was able to make a good backup. But I couldn’t recover. So then, from this hole I was in, I started to dig myself in farther, aiming (so it seems) for the center of the earth.
Continue reading Another self inflicted wound
I don’t know if I actually know enough to write this post. But I want to record what little I do know about this.
The symptom is that my tarragondata volume on this system, tarragon, claims to be out of space. This is a btrfs volume, about which there are other posts. It contains most of the dynamic parts of the system. The root volume ‘/’ is very small, about 20GB. Just enough to install the Centos code and keep a few little things. The great majority of the information needed to run the system is symlinked out of /, which includes /home, mail, databases, websites and their data, the repositories, certificates, local scripts etc.
This is a 180GB disk, and it currently is running about 55% full, i.e. almost 100GB used. Among the information on this disk are snapshots of all the tarragondata, every night for 30 days. This isn’t disaster backup/disk failure backup (which is elsewhere), this is “operator error” backup.
A couple of weeks ago I began to experience a new kind of failure. In the middle of the night, suddenly this btrfs volume would report that it was out of space – usage 100%, although the amount of storage in use was, still the roughtly 100GB that it normally uses. It manifestly was not actually out of space.
Continue reading Out of space on btrfs
I’m an old guy, and I mostly use pretty old stuff, but occasionally I think I should migrate to more up to date tools. I’ve used svn for 20 years. I know I should have moved to git long ago. I have also been using Eclipse as an IDE since forever. I run on apache. My systems aren’t virtual. One of my main websites is still hand wrought in PHP, and another is using an ancient and obsolete version of zend framework which has been abandoned by Zend. I think of moving it, maybe to laravel, maybe even to ruby on rails.
This documents my effort to modernize, if only a little. This is a very uncomfortable process because there are times when I’m sure I’m going to foul up the running sites, and be unable to get them up, or that I’m not going to be able to fix the next bug that comes along.
Continue reading The working environment