Timeshift is the cache machine.
Forum rules
Do not post support questions here. Before you post read the forum rules. Topics in this forum are automatically closed 6 months after creation.
Do not post support questions here. Before you post read the forum rules. Topics in this forum are automatically closed 6 months after creation.
- Pjotr
- Level 23
- Posts: 19802
- Joined: Mon Mar 07, 2011 10:18 am
- Location: The Netherlands (Holland) 🇳🇱
- Contact:
Re: Timeshift is the cache machine.
Nasty. Anyway, for me this is yet another reason not to automate the creation of Timeshift snapshots.
After all, practically nobody needs to have more than two snapshots for repairing a broken system. One at least a month older than the other (in order to decrease the risk that they both contain the same "heritage" of errors).
No matter how old a restored snapshot is: simply run Update Manager and you'll be up to date in no time at all. For the average user, all this intensive snapshot-taking is pure baloney, in my opinion. Occupational therapy for the bored.
After all, practically nobody needs to have more than two snapshots for repairing a broken system. One at least a month older than the other (in order to decrease the risk that they both contain the same "heritage" of errors).
No matter how old a restored snapshot is: simply run Update Manager and you'll be up to date in no time at all. For the average user, all this intensive snapshot-taking is pure baloney, in my opinion. Occupational therapy for the bored.
Tip: 10 things to do after installing Linux Mint 21.3 Virginia
Keep your Linux Mint healthy: Avoid these 10 fatal mistakes
Twitter: twitter.com/easylinuxtips
All in all, horse sense simply makes sense.
Keep your Linux Mint healthy: Avoid these 10 fatal mistakes
Twitter: twitter.com/easylinuxtips
All in all, horse sense simply makes sense.
- antikythera
- Level 15
- Posts: 5723
- Joined: Thu Jul 02, 2020 12:52 pm
- Location: Cymru
Re: Timeshift is the cache machine.
Timeshift is not meant to send snapshots to more than one drive so it's no surprise it acts oddly when you do so.
I’ll tell you a DNS joke but be advised, it could take up to 24 hours for everyone to get it.
Re: Timeshift is the cache machine.
Some more details would help to understand, what happens here.
This happens after which number of snapshots?
The provided information is not even enough to find out "if anyone else has ever noticed this".
Which cache?
You mean, that you do manually snapshots and switch for each snapshot the target partition? Which user does this? (Leaving open, what happens really.)
What gets written to disk? Where does it get written? To swap? To somewhere else?
This happens after which number of snapshots?
The provided information is not even enough to find out "if anyone else has ever noticed this".
- antikythera
- Level 15
- Posts: 5723
- Joined: Thu Jul 02, 2020 12:52 pm
- Location: Cymru
Re: Timeshift is the cache machine.
I've not noticed it happen, I don't have automatic snapshots enabled for mintupdate and have timeshift limited to 2xdaily and 2xboot snapshots only.
I use foxclone to image the drives in my linux machines once a month to 1x NAS and 2x removable media drives in rotation. The latter get locked in a fire and flood resistant cabinet.
Timeshift is only there to rollback a bad decision by me or an update that causes issues, therefore I don't need loads of snapshots over different drives.
I use foxclone to image the drives in my linux machines once a month to 1x NAS and 2x removable media drives in rotation. The latter get locked in a fire and flood resistant cabinet.
Timeshift is only there to rollback a bad decision by me or an update that causes issues, therefore I don't need loads of snapshots over different drives.
I’ll tell you a DNS joke but be advised, it could take up to 24 hours for everyone to get it.
- antikythera
- Level 15
- Posts: 5723
- Joined: Thu Jul 02, 2020 12:52 pm
- Location: Cymru
Re: Timeshift is the cache machine.
No doubt it would
I’ll tell you a DNS joke but be advised, it could take up to 24 hours for everyone to get it.
Re: Timeshift is the cache machine.
What has automatic snapshots to do with it? If the OP uses "several drives" (that would mean at minimum 2, than there cannot be any automation in the game, because TS does not use several drives, as you noted yourself already.antikythera wrote: ⤴Sat Jul 02, 2022 6:28 pm I don't have automatic snapshots enabled for mintupdate
The shown result is meaningless. The OP wrote:
I have one drive for my snapshots and I see the same amount of cached data as in the shown picture (actually a little bit more). So the claim, that using the cache is the result of using several drives for snapshots is wrong, regardless what gets asserted.
As no help is requested the starting post has already the purpose to claim something; but what has been claimed contradicts the most simple observations.
- smurphos
- Level 18
- Posts: 8501
- Joined: Fri Sep 05, 2014 12:18 am
- Location: Irish Brit in Portugal
- Contact:
Re: Timeshift is the cache machine.
How do you know it's timeshift and why do you think it's a problem? It's normal for Linux to fill the page cache in use- it's just data that has been read from storage for whatever reason and sits in RAM in case its needed again but is instantly reclaimable should the system need it for something else.
For custom Nemo actions, useful scripts for the Cinnamon desktop, and Cinnamox themes visit my Github pages.
Re: Timeshift is the cache machine.
There is no chance to verify anything, if no precise information and answers are given.
How determine? Precisely described!
The problem is, that you simply repeat your claims, but you give no proof. If it would be "very easy to determine" than even an idiot like me should be able to determine this. Fact is, that I determine nothing, what you write.
No, you describe your conclusions out of an detail, but you give no proof, that the conclusion is founded.
I asked you what gets written. No answer.
You wrote:
Even assumed, this would be correct: cache is not trash. So what it is about? You even contradict yourself, in the picture by you there are only 5 GB (1/3) cache displayed.
-
- Level 1
- Posts: 9
- Joined: Thu May 13, 2021 7:31 am
- Location: Japan
Re: Timeshift is the cache machine.
Hello, I was also suffering from the same event.
(It certainly doesn't do any real harm)
I have noticed a sudden increase in OS cache usage in the past, which is when the weekly backups by Timeshift went live. Just today was the day the weekly backup was to go live, so I tried to get data before and after the operation.
This is the uptime result before operation. I am not rebooting the PC and am running it in suspend.
This is a list of weekly backup files by timeshift before going live. The last time it was on August 16.
And this is the virtual memory usage. swap usage(7.0Mi) occurred on August 16. Also, the drop-cache command was submitted after that to free up the OS cache. At that time, OS cache usage was about 10Gi.
This is the results of the uptime command after the weekly backup is up and running.
This is a list of weekly backup files by timeshift after going live. The time is updated on August 23.
And this is the virtual memory usage. buffers has increased from 697Mi to 1.0Gi and cache has increased from 4.5Gi to 5.8Gi. The increase is less than the last weekly backup, but I am sure that the increase is due to the timing of the timeshift operation.
I assume that this is due to the rsync command issued internally by timeshift.
I think that replacing the rsync command with (nocache rsync ~) may solve this event.
Sorry for my poor English.
(It certainly doesn't do any real harm)
I have noticed a sudden increase in OS cache usage in the past, which is when the weekly backups by Timeshift went live. Just today was the day the weekly backup was to go live, so I tried to get data before and after the operation.
This is the uptime result before operation. I am not rebooting the PC and am running it in suspend.
Code: Select all
$ uptime
09:46:27 up 7 days, 16:10, 1 user, load average: 0.47, 0.78, 0.89
Code: Select all
$ ll -h
合計 8.0K
drwxr-xr-x 2 root root 4.0K 8月 16 10:01 ./
drwxr-xr-x 9 root root 4.0K 8月 16 10:01 ../
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-06-12_22-00-01 -> ../snapshots/2022-06-12_22-00-01
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-06-20_06-00-01 -> ../snapshots/2022-06-20_06-00-01/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-06-27_08-00-01 -> ../snapshots/2022-06-27_08-00-01/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-07-04_08-00-01 -> ../snapshots/2022-07-04_08-00-01/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-07-11_08-00-01 -> ../snapshots/2022-07-11_08-00-01/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-07-15_09-00-02 -> ../snapshots/2022-07-15_09-00-02/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-07-22_09-00-02 -> ../snapshots/2022-07-22_09-00-02/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-07-29_09-00-01 -> ../snapshots/2022-07-29_09-00-01/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-08-02_10-00-02 -> ../snapshots/2022-08-02_10-00-02/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-08-09_10-00-01 -> ../snapshots/2022-08-09_10-00-01/
lrwxrwxrwx 1 root root 32 8月 16 10:01 2022-08-16_10-00-01 -> ../snapshots/2022-08-16_10-00-01/
Code: Select all
$ free -hw
total used free shared buffers cache available
Mem: 15Gi 3.9Gi 6.4Gi 40Mi 697Mi 4.5Gi 11Gi
Swap: 7.8Gi 7.0Mi 7.8Gi
This is the results of the uptime command after the weekly backup is up and running.
Code: Select all
$ uptime
10:00:55 up 7 days, 16:25, 1 user, load average: 0.74, 0.81, 0.91
Code: Select all
$ ll -h
合計 8.0K
drwxr-xr-x 2 root root 4.0K 8月 23 10:00 ./
drwxr-xr-x 9 root root 4.0K 8月 23 10:00 ../
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-06-20_06-00-01 -> ../snapshots/2022-06-20_06-00-01/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-06-27_08-00-01 -> ../snapshots/2022-06-27_08-00-01/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-07-04_08-00-01 -> ../snapshots/2022-07-04_08-00-01/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-07-11_08-00-01 -> ../snapshots/2022-07-11_08-00-01/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-07-15_09-00-02 -> ../snapshots/2022-07-15_09-00-02/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-07-22_09-00-02 -> ../snapshots/2022-07-22_09-00-02/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-07-29_09-00-01 -> ../snapshots/2022-07-29_09-00-01/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-08-02_10-00-02 -> ../snapshots/2022-08-02_10-00-02/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-08-09_10-00-01 -> ../snapshots/2022-08-09_10-00-01/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-08-16_10-00-01 -> ../snapshots/2022-08-16_10-00-01/
lrwxrwxrwx 1 root root 32 8月 23 10:00 2022-08-23_10-00-01 -> ../snapshots/2022-08-23_10-00-01/
Code: Select all
$ free -hw
total used free shared buffers cache available
Mem: 15Gi 3.9Gi 4.9Gi 45Mi 1.0Gi 5.8Gi 11Gi
Swap: 7.8Gi 7.0Mi 7.8Gi
I assume that this is due to the rsync command issued internally by timeshift.
Code: Select all
[10:00:01] RsyncTask:execute()
[10:00:01] export LC_ALL=C.UTF-8
rsync -aii --recursive --verbose --delete --force --stats --sparse --delete-excluded --link-dest='/run/timeshift/965422/backup/timeshift/snapshots/2022-08-16_10-00-01/localhost/' --log-file='/run/timeshift/965422/backup/timeshift/snapshots/2022-08-23_10-00-01/rsync-log' --exclude-from='/run/timeshift/965422/backup/timeshift/snapshots/2022-08-23_10-00-01/exclude.list' --delete-excluded '/' '/run/timeshift/965422/backup/timeshift/snapshots/2022-08-23_10-00-01/localhost/'
[10:00:01] RsyncTask:prepare(): saved: /tmp/3boZgdut/2022-08-23_10-00-01/script.sh
[10:00:01] AsyncTask: child_pid: 965456
[10:00:28] AsyncTask: finish(): enter
[10:00:28] exit_code: 0
Sorry for my poor English.