Woops… Maybe I need to be a sticker for detail. I missed the M vs G. Will fix. Thank you for pointing that out. I helped family move and was quite out of energy when I worked on this. Will pay closer attention.
2 Likes
It might be time to consider a larger SSD – at least 500 GiB – though if you’re very careful, and depending upon your use case, you might be able to make your existing 250 GiB workable. 
(I personally prefer not to use anything below 1 TB, but definitely not below 500 GiB for an OS).
OK… I have the swap file fixed and instead of just going to 47G, I went up to 68G
free -h
total used free shared buff/cache available
Mem: 62Gi 3.2Gi 58Gi 107Mi 1.7Gi 59Gi
Swap: 68Gi 0B 68Gi
My SSD that Manjaro is installed on is 2TB so plenty of room left.
I will let you know if this seems to solve the system freezes, and I will be booting the 6.18 LTS kernel.
uname -r
6.18.12-1-MANJARO
Thank for all the help.
4 Likes
Found the system frozen this morning. Had Firefox running, though not really doing anything and Dolphin running, but also not really doing anything.
The REISUB procedure did not work, could not get to a TTY, etc.
Booted into a live USB to run fsck and make sure file system was OK. Everything was fine.
Guess I will go back to Kernel 6.1 for now as it has been the most stable. I have had 6.6, 6.12 and 6.18 all cause hard freezes. 6.1 hiccups but always seems to come back, though the performance (when it is working) is so much better on the newer kernels.
Will keep an eye on the AMD glitch/race bug that is out there, and start learning more about digging thru the logs that are generated (though I tried this about a month ago and don’t think anything is making it into the logs when the system freezes).
If anyone else has any other thoughts or suggestions, they are appreciated.
One other thought, can I run other kernels like the zen kernel, or is that just a really bad idea on Manjaro?
Thank you.
Did you remember to activate the swap partition?
That kernels are the cause of your issue is highly unlikely, especially considering (as you say) the crashes/freezes occur regardless of the kernel in use.
Could you switch back to Wayland for a few days and see if there is any difference. I suggest this only in the spirit of “trial and error”.
1 Like
Logs.
1 Like
Possibly a stupid question, but you did actually enable it beforehand as per the instructions?
1 Like
The DDR5 seems to run outside normal specs cf. DDR5 SDRAM - Wikipedia Specs are between 4000 MT/s and 8800 MT/s. The actual value is shown as 3600 MT/s. What is the reason for this?
1 Like
Most likely half a channel is filled. Memory modules usually need to be installed in pairs. With one, It either works at half speed, or not at all. So the former seems likely.
The good news, if that’s the case, is that one 16GB DIMM will double the size and speed of your memory.
I jumped the gun seeing it at half the supposed speed.
Memory:
System RAM: total: 64 GiB available: 62.44 GiB used: 4.17 GiB (6.7%)
Message: For most reliable report, use superuser + dmidecode.
Array-1: capacity: 128 GiB slots: 4 modules: 4 EC: None
max-module-size: 32 GiB note: est.
Device-1: Channel-A DIMM 0 type: DDR5 detail: synchronous unbuffered
(unregistered) size: 16 GiB speed: spec: 4800 MT/s actual: 3600 MT/s
volts: note: check curr: 1 min: 1 max: 1 width (bits): data: 64 total: 64
manufacturer: Corsair part-no: CMH32GX5M2N6400C36 serial: N/A
Device-2: Channel-A DIMM 1 type: DDR5 detail: synchronous unbuffered
(unregistered) size: 16 GiB speed: spec: 4800 MT/s actual: 3600 MT/s
volts: note: check curr: 1 min: 1 max: 1 width (bits): data: 64 total: 64
manufacturer: Corsair part-no: CMH32GX5M2N6400C36 serial: N/A
Device-3: Channel-B DIMM 0 type: DDR5 detail: synchronous unbuffered
(unregistered) size: 16 GiB speed: spec: 4800 MT/s actual: 3600 MT/s
volts: note: check curr: 1 min: 1 max: 1 width (bits): data: 64 total: 64
manufacturer: Corsair part-no: CMH32GX5M2N6400C36 serial: N/A
Device-4: Channel-B DIMM 1 type: DDR5 detail: synchronous unbuffered
(unregistered) size: 16 GiB speed: spec: 4800 MT/s actual: 3600 MT/s
volts: note: check curr: 1 min: 1 max: 1 width (bits): data: 64 total: 64
manufacturer: Corsair part-no: CMH32GX5M2N6400C36 serial: N/A
This is fine, it’s one of many DDR5 standards @ 4800 MT/s. Why it’s slightly lower I do not know.
1 Like
Possibly underclocked. BIOS configuration. Modules need re-seating. Not the ideal RAM for the board (if built themselves). Bad RAM. Mainboard is “too new”. 
Hard freezing with no log entries is always suspicious for hardware problem. Even if it was memory it would have logged OOM.
1 Like
Yes, I activated the REISUB capability and even pre-tested it prior to the latest freeze so I knew it would work.
echo $XDG_SESSION_TYPE
wayland
swapon --show
NAME TYPE SIZE USED PRIO
/dev/nvme0n1p3 partition 68.4G 0B -2
2 Likes
I asked for logs twice, and how to get them if oe86 does not know how.
Or did I miss something, and there are no logs?
1 Like
Sorry Molski. I should have responded to this. Its been a very busy day. I will look into the logs most likely tomorrow and will provide ASAP.
2 Likes
Kernel 6.1 was originally release December 2022 - around the same time your CPU was released.
The kernel continually improve and that is especially true for AMD based systems - so expect the latest kernel to be the one having the best support for your system.
You can easily build a kernel tailored for your system.
Can the "how-to-compile-the-mainline-kernel-the-manjaro-way" be revived? - #15 by linux-aarhus
By using the above recipe you can maintain a kernel that is only changed when you choose to recompile the kernel.
You will likely need to experiment to experiment to get it right but the knowledge and experience gained is well worth the effort.
3 Likes
first , update your UEFI mothertboard
redo all options after reset
for memory DDR5 , with 4 slots PCI , you CANT uses the with same latency as 2 Slots Latency , can you provide informations on this ? ( ref model memory )
So after the 2026-03-23 stable update, I was still having hard freezes while running the 6.18 LTS kernel, but I noticed the update included a new 7.0 RC kernel which I installed. I have been running solid now for 7 days straight.
Prior to this, I would hard freeze once a day, so I think this may be resolved in the 7.0 RC kernel.
I also should have mentioned I have other distros installed on this machine (endeavour, Pop, Fedora and Mint). Those distros did not have the hard freeze problem, so I did not think this was hardware related, but “you don’t know until you know”.
In either case, I have not had much time to spend on the computer, other than setting it up to try things like the new kernel, other distros and then just sitting there running to see if I would come back and find it locked up. Have had a ton of family challenges going on, and family comes before computers.
Thanks for the help and I think this may be resolved with 7.0RC
Will see what happens as more time goes by, but I really like being able to rely on Manjaro again and not booting in to other distros. Manjaro is 100% my preference.
Still plan on digging into logs but just too much going on right now.
This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.