205 reputation
19
bio website
location
age
visits member for 2 years, 2 months
seen Mar 3 at 22:23

Mar
3
comment How long does it usually take to format a 2TB external HD using ext4 and gparted?
ext4 can be formatted in a defered way, where most of formatting commences in background while the filesystem can already be used after a very short initalization period. Using mkfs.ext4, the option would be -E lazy_itable_init=1, and this even is a default setting it seems. I don't know how to make gparted using this option, but maybe it is possible. Using this feature, a 2GB partition gets available after about one minute.
Oct
17
comment rdiff-backup over existing backup
Sounds good! As I am not using rdiff-backup in daily procedure now, may I ask if you tried this successfully?
Sep
13
comment Is it possible to make the OOM killer intervent earlier?
I think this answer is a bit like 'don't do anything real bad to solve a problem' but in my opinion a system that can made unresponsive by a bunch of seemingly harmless user actions is worse than a system that kills them early.
Aug
28
comment How to store file attributes like permissions and timestamps and reapply them later on linux?
Yes, but I want to store the attributes alone. I use other tools to store the files, but they do not reliable store attributes. So I like to use something that is able to store attributes but don't neccessary the file contents.
Aug
25
comment How to sync file times from one directory tree to another
That seems to be a dead end. rsync works well if the trees are identical except the timestamps, but if there are any other changes, it gets too complicated to keep rsync from applying them either.
Aug
25
comment How to sync file times from one directory tree to another
This works well, except it is a little slow as you stated. Suprisingly, it does makes several hundred files per second, which is quite cool if you remember a new touch is run for every file.. however, I will try a Python script like you propose.
Aug
25
comment How to sync file times from one directory tree to another
Even when using --existing, there is still some risk loosing modifications made to some files since the duplication. Maybe there is some way preventing rsync from copying content at all? Then the problem would shrink to some resetted timestamps. On the other hand, using rsync without --size-only would give a precise picture of what files would need a new timestamp and which do not. Using with -n for dry run, I would get a list stating >f for files that would be reseted but should not, and .f.t for files that are old and would get the correct old timestamp. But how to apply it then?
Aug
25
comment How to sync file times from one directory tree to another
Looks cool. I also added --existing to prevent rsync copying files again that where deleted after the duplication.
Aug
24
comment cpu load measure with hyperthreading on linux
Simply spoken: How to tell at a given moment, if my CPU would be capable to do further work, without slowing down the currently ongoing work?
Aug
24
comment cpu load measure with hyperthreading on linux
My question is how to get the total usage of available cpu cycles. I think you're saying 'you can't get that value because the OS isn't capable of knowing it', but I am not aware why. The OS may know if hyperthreading happens by probing the behaviour or having some CPU driver defining if the current CPU model uses it or not. In worst case, the OS truely is incapable of handling this, I like to have a suggestion how to compute the value by myself using my knowledge of my current CPU hyperthreading capabilities.
Jul
25
comment cpu load measure with hyperthreading on linux
This is obviously a reason why the OS should be aware of virtual cores. So it has to find out and compute the physical core usage based on it. Otherwise the whole concept of measureing "usage" and "load" has no use to the user. If I run top, I usually have some question like "Is the system running at it's limit?" or "Would it be useful to divide the work into more processes?" etc. This questions can't be reliable answered by the current topoutput.
Jul
19
comment cpu load measure with hyperthreading on linux
I just like to have a load measure where 100% would mean that every cycle of every real core is used.
Jul
19
comment cpu load measure with hyperthreading on linux
It seems so, but it doesn't have to? The real vs. virtual core mapping is a simple one to two map. The problem is how to measure load on a virtual core that actually changes its available performance by getting scheduled with another one on the real core. But all data is avalable I think, the question is just where are the tools that get a proper result out of them?
Jun
28
comment Fast and reliable reboot after power loss
This can also be combined with another eg. ext3 read-writable partition where single databases or log files are stored. As this partition is usually small and low on file count, it can be fscked in a short time and is unlikely to fail completely. Even if data gets corrupted, the system boots up, and your applications can be resilent to missing data if well made.
Mar
10
comment How to make Windows 7 switch between programs faster after a while?
As another example, usual Linux distributions do not swap out anything until the RAM is almost full, which seems to avoid this problem at all. Maybe Windows could be switched to this behaviour too.
Feb
16
comment Fast and reliable reboot after power loss
The LiveCD was a good hint.. actually it seems a readonly partition overlayed with a writable tempfs would do the job for 'kiosk'-only uses. The harddisk would be mounted readonly this way, suffering no effects from powering off. So the fs should stay clean and reboot like usual every time. Will try this out.
Feb
16
comment Fast and reliable reboot after power loss
A rock solid solution for presentation applications would be having the harddisk mounted read only, and use OverlayFS or something else to get a read write capable root fs, that only writes to a temporary ram fs. However it seems quite complex to turn a ordinary distribution into such a setup.
Apr
5
comment Is it possible to make the OOM killer intervent earlier?
@TomWijsman: I don't like introducing anything into the application because this is a general problem that can be caused by several applications. That's why I still see the OOM as the apropriate tool. ulimit -v is unrealiable because it can only limit one application and doesn't care about the total available ressources.
Mar
29
comment Is it possible to make the OOM killer intervent earlier?
I have no problems with killing applications that would run dead either. Consider a system with 2GB physical + 2GB swap. An application that quickly runs out the physical memory can easily eat the swap too. It would just die later, after rendering the system unresponsive for minutes to hours. So why not kill it quickly before GUI operation get flaky? Much processes do all their work with 10mb, some take 1gb, and some rare would need 10gb, that's life.
Mar
29
comment Make Chrome browser prefer DNS over searching
yes it is. Thank you.