The small linux problem thread

[94674.908396] usb 1-4: new high-speed USB device number 5 using xhci_hcd
[94675.119444] usb 1-4: New USB device found, idVendor=0781, idProduct=5530
[94675.119448] usb 1-4: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[94675.119451] usb 1-4: Product: Cruzer
[94675.119453] usb 1-4: Manufacturer: SanDisk
[94675.119455] usb 1-4: SerialNumber: 2005173841168FB2FC8B
[94675.123499] usb-storage 1-4:1.0: USB Mass Storage device detected
[94675.124124] scsi host14: usb-storage 1-4:1.0
[94676.154449] scsi 14:0:0:0: Direct-Access     SanDisk  Cruzer           1.20 PQ: 0 ANSI: 5
[94676.155140] sd 14:0:0:0: Attached scsi generic sg5 type 0
[94676.155842] sd 14:0:0:0: [sdf] 7821312 512-byte logical blocks: (4.00 GB/3.73 GiB)
[94676.157653] sd 14:0:0:0: [sdf] Write Protect is off
[94676.157656] sd 14:0:0:0: [sdf] Mode Sense: 43 00 00 00
[94676.159831] sd 14:0:0:0: [sdf] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
[94676.170139]  sdf: sdf1
[94676.174662] sd 14:0:0:0: [sdf] Attached SCSI removable disk

I get this on fedora after plugging in my flash drive. nothing additional after running blkid.

Thereā€™s nothing wrong with systemd, except bloat. Itā€™s very good at what it does and Iā€™m leaving it at that. Also, I can almost guarantee itā€™s unrelated to your issue.

Okay, this mght be a stupid question, but is there a reason you need access to the SD card reader? I might be misunderstanding, but I thought this Pi was acting simply as a gateway.

It does act as gateway/router/NAT, but I also want a small cloud system in it (mostly for testing stuff), and for that some additional storage is kinda needed :slight_smile:
The microSD card in the Pi is 32GB and extending that with a 64GB microSD is just great for extra things I might want to test.

For the time being I wrote myself a script that runs ā€œblkidā€ a few times when it detects that the stick is plugged in and then mounts the partition. Though that is not an optimal solution at allā€¦

1 Like

Yeah, I canā€™t say whatā€™s going on there. Itā€™s probably a shitty reader.

1 Like

This isnā€™t exactly a problem, more of a question.
What is the state of TPM passthrough, is it usable enough to be used on a work laptop?
I have a colleague who has had issues with it in the past and the QEMU wiki page was last updated in 2012 https://wiki.qemu.org/Features/TPM

Why does my screen/fonts scale differently when using nvidia drivers as opposed to nouveau? I like the look of nouveau better, but games run better on nvidia.

I use the same xrandr cmd line for screen placement, just the ports are named different.

xrandr --output HDMI-1 --pos 0x180 --output DVI-D-1 --pos 1920x0 
xrandr --output HDMI-0 --pos 0x180 --output DVI-D-0 --pos 1920x0

I have not changed another setting as far as Iā€™m aware or it should be hiding somewhere in my ~.

I would look up the chipset of the stick and look for known issues. More than likely, there is something going on and the stick is not in the correct mode at start up. The block id command probably forces and additional mode.

Eitherway, since you are on OpenRC, there is probably something that it may not have privileged access to at the time or something that the scripts are missing. Since arch is bleeding edge, it may have workarounds built in. Check the Arch Wiki

The LTE stick needed flashing at the beginning (when I got it, becacause it only had a webinterface and did not expose ttys) to expose the NCM interface and go into ā€œstick modeā€, it works flawlessly in Windows or Arch Linux with systemd.
Just to give some more info, the system was running (on-demand) for almost 8 months with Arch and it never was an issue.

OpenRC is in conjunctional use with Alpine Linux where the problem is actually a thing (tested on another server machine running Alpine, same problem).

@Hako
That proves my point. Systemd runs as PID 1. It is omnipotent. Open RC is designed to be a replacement for SystemV but again held together with scripts that are not so integrated into the core of the system like Systemd. If it works in Arch, use the wiki to figure out why it works the way it does. Then try to back port that to Alpine and Open RC. If you are not up to the task, I would say stick with arch. It is very possible that you are the only one having this issue with Alpine because you are the only one that is doing it. You are going to need to do some research to properly resolve the issue. Otherwise, your solution that you have in place now will have to do.

I would start looking here as I am sure this is where your issue lies.


ran into this issue, havenā€™t updated my laptop for a while

note to self if db lock issues occur rm db.lck to get it to update

 sudo rm /var/lib/pacman/db.lck
1 Like

Installed Linux Mint 19 this weekend.
And I canā€™t find how to make it resume from suspend via keyboard or mouse, both USB.
The keyboard has power, it turns on its little screen when I press any key while the computer is suspended.
Any hints?

Network discovery seems to be the main issue behind my earlier trouble with samba (still unsolved).

Samba works by specifying the shareā€™s host IP in a file manager or smbclient. But browsing through the ā€œNetworkā€ entry in file managers brings up nothing. Itā€™s not just Fedora 28. I also tried Ubuntu live USB 18.04 and Manjaro. Same behavior in all of them.

Solus has a long thread on the subject and they apparently solved it. https://dev.solus-project.com/T1223

Tips on recovering network discovery?

Iā€™ve had this issue for quite a long while now. Itā€™s more of a nuisance than anything else, but Iā€™m going on vacation for a week and Iā€™d like to be able to use my laptop reliably.

I have an entry in my fstab to mount my fileserver nfs share at boot. When I leave my network and my fileserver canā€™t be seen file explorer (Gnome Files or whatever itā€™s called now) locks up when navigating to my home folder, which is where the nfs share is mounted. Eventually it must fail enough times that it gives up and I can access my home folder again.

Is there an option I can add to the fstab line to make this stop? Iā€™m not at home, so I canā€™t post my fstab right now. But if itā€™ll help I can do that this evening.

This is a quote from the arch wiki and what I used-

Remote filesystem

The same applies to remote filesystem mounts. If you want them to be mounted only upon access, you will need to use the noauto,x-systemd.automount parameters. In addition, you can use the x-systemd.device-timeout= option to specify how long systemd should wait for the filesystem to show up. Also, the _netdev option ensures systemd understands that the mount is network dependent and order it after the network is online.
noauto,x-systemd.automount,x-systemd.device-timeout=30,_netdev

wiki.archlinux.org/index.php/fstab

1 Like

Great user name haha!

1 Like

The important option in the quote that @TheCakeIsNaOH mentioned is going to be noauto.

if you pass that option, any init system will not automatically mount the filesystem.

The problem with NFS is that itā€™s a really dumb (read simple) protocol. It doesnā€™t really handle inconsistent (or lost) connections very well, so when you drop off the network, it still thinks itā€™s connected. If you post a bit more details about your config, I could definitely try to get you a better solution.

I think your use case is the reason the soft option exists, although it is ā€œriskyā€.

1 Like

The line for the nfs share in my fstab is this:

192.168.0.248:/mnt/homeserver_share /home/marasm/homeserver nfs rw,soft,retrans=6,timeo=60,intr,rsize=8192,wsize=8192,timeo=14 0 0

Idk if soft and intr can work together, but not 100%. I honestly usually use ā€œautoā€ unless I need to eke out performance.

Are the two timeo options an issue? Donā€™t know why thereā€™s two.

Oh I just noticed that. Yeah I donā€™t think any parameter can be in there twice.

Did you come up with those or was it like a copy/paste situation?

ā€”-

Try this out:

192.168.0.248:/mnt/homeserver_share /home/marasm/homeserver nfs rw,retrans=2,timeo=60,intr,rsize=8192,wsize=8192 0 0

That should yield a 12 second hang of the share isnā€™t available. Reduce timeo or retrans if thatā€™s too long.