View previous topic :: View next topic |
Author |
Message |
MMMMM Tux's lil' helper
Joined: 13 Jun 2011 Posts: 141 Location: Berlin
|
Posted: Sun Nov 26, 2023 4:34 pm Post subject: [solved] "Performing Global Updates" takes forever |
|
|
Hi,
when doing emerge --sync, sometimes emerge does at the end a "Performing Global Updates" task. It takes normally some minutes. This time it took like 6 or 8 hours, then I cancelled it:
Code: |
Performing Global Updates
(Could take a couple of minutes if you have a lot of binary packages.)
.='update pass' *='binary update' #='/var/db update' @='/var/db move'
s='/var/db SLOT move' %='binary move' S='binary SLOT move'
p='update /etc/portage/package.*'
/mnt/portage/portdir/profiles/updates/4Q-2023...................................
@%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%@%%%%%%%%%%%%%%%%%%%%%%%%%%%%%@%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%sSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS@%%%%%%%%%%%%%%%%%%%%%%%%%%%%%#####################################################################################################################################################################################################################
#################################################################################################################################**************************************************************************************************************************************************
****************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
****************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
*****************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
******************************************************************************************************************************************************************************************************************************************************************************************************************************************************************
**************************************************************************************************************************^C
Exiting on signal 2
|
Restart seems really restart instead of doing it again. How do I find out how much is already done?
*****
Solution: reduced PKGDIR from 150 GB to 10 GB, "Performing Global Updates" then took only a couple of minutes.
[Moderator edit: broke long whitespace-free lines in code tags to fix thread layout. -Hu]
Last edited by MMMMM on Sun Nov 26, 2023 7:45 pm; edited 1 time in total |
|
Back to top |
|
|
sam_ Developer
Joined: 14 Aug 2020 Posts: 1938
|
Posted: Sun Nov 26, 2023 4:54 pm Post subject: |
|
|
I assume you have some binpkgs. Do you have a lot of them? Is PKGDIR on something like NFS or slow media? |
|
Back to top |
|
|
MMMMM Tux's lil' helper
Joined: 13 Jun 2011 Posts: 141 Location: Berlin
|
Posted: Sun Nov 26, 2023 6:08 pm Post subject: |
|
|
It's on NFS.
Approx. 150 GB
Code: | /mnt/pkgdir # du -sm .
152020 . |
I have plenty of space left on an internal HDD. But not on the SSD.
I guess moving this to the HDD would be a good idea. |
|
Back to top |
|
|
Hu Administrator
Joined: 06 Mar 2007 Posts: 22578
|
Posted: Sun Nov 26, 2023 6:24 pm Post subject: |
|
|
Do you need all 150GB of those packages? If you clean out old packages, Portage will have less work to do. If you can run Portage directly on the server, letting it process the moves there (where the files will be local) may avoid needing the client to process them over NFS. I have never encountered this situation, so I cannot say whether processing the moves server-side will help the client, but it seems like it is worth a try. |
|
Back to top |
|
|
Goverp Advocate
Joined: 07 Mar 2007 Posts: 2167
|
Posted: Sun Nov 26, 2023 7:13 pm Post subject: |
|
|
MMMMM,
before your next sync will dispose of all binary packages except those you currently have installed. My system, with around 1,600 packages, 250 in @world, needs about 7GB of binpkgs. Doing it before syncing allows you to fall back if an update turns nasty. If you're desperate for old versions, omit the "--deep", but you'll need more space, and even then, some packages will have only one or two versions available. _________________ Greybeard |
|
Back to top |
|
|
MMMMM Tux's lil' helper
Joined: 13 Jun 2011 Posts: 141 Location: Berlin
|
Posted: Sun Nov 26, 2023 7:41 pm Post subject: |
|
|
I copied all to the internal HDD. Was not that much faster. After eclean-pkg only 10 GB left, emerge --sync then finished quickly (some minutes).
I have that in my mainenance script. But disabled; can't remember why. |
|
Back to top |
|
|
Plus! n00b
Joined: 04 Mar 2008 Posts: 15
|
Posted: Thu Mar 28, 2024 11:31 pm Post subject: |
|
|
I have the similar problem. Package dir is on NFS share, however the size is about 18GB.
And the global update (*='binary update') takes hours. |
|
Back to top |
|
|
|