"2321" packages to upgrade . . . on reboot another "204" yet to go?? Why???

English Open Chat
121

@karlm:

Thanks for that . . . I do know there is a period of time where packagekit would be blocking manual zypper-ing . . . it’s difficult to know when all of that overlapping is happening . . . .

For now, as per my last post, unless you have some response to the question about whether shutting down while possibly in the middle of an unseen “dupa.service” would have no effects . . . ??? I have “stopped” the services for the time in which we just have the “nominal” package upgrades . . . and then I can “start” them again if and when the tsunami returns . . . .

Would I have to run “systemctl daemon-reload” after starting them up or before, or not at all, should be easy to re-fire the local services and then let the machine run for a longer period of time until we might figure dupa had processed the packages??

In this case I can’t say “12 minutes of cpu time” for 378 packages is “super-quick”??? That seemed to be over 40 minutes via clock time . . . .

122

Clock time depends on unknowns, Fragments of information are useless. Always provide a complete set.

I upgraded a budget notebook (440 €) this morning:

Operating System: openSUSE Tumbleweed 20220604
KDE Plasma Version: 5.24.5
KDE Frameworks Version: 5.94.0
Qt Version: 5.15.2
Kernel Version: 5.18.1-1-default (64-bit)
Graphics Platform: X11
Processors: 8 × Intel® Core™ i5-8250U CPU @ 1.60GHz
Memory: 7.7 GiB of RAM
Graphics Processor: Mesa Intel® UHD Graphics 620

Tumbleweed 20220414-0 -> 20220604-0, 2442 packages upgraded, 40 new, 28 deleted, total size download: 1,91 GiB. Consumed 8min 46.395s CPU time. Install consumed 13min 57.110s CPU time.

123

OK . . . “fragments” . . . well this morning there was “3886 lines” of dupa processing for the 378 packages showing in the journalctl report . . . is that the “complete set” you would want to see???

The more relevant question is on the blithe shutting down of the machine as I have to break and run to get to work . . . and possibly dup/dupa could have run some packages and then later on in the session found some more and would start running them w/o announcing that to me the “pilot” . . . but I would need to shut the machine down???

Would that interruption of dupa process cause potentially problematic outcomes?? Or, no, even if I shut down while dupa is running, it will just re-start on booting TW???

From what I saw in journalctl it looks like “dupa” shows both the dup download and then runs the remove/install of the packages . . . which didn’t exactly fly by . . . .

For now I have two or three more TW systems to install the service, but in TW I have “stopped” it for now . . . since I come and go in and out of the system in a few hours.

124

What I posted in #122.

The more relevant question

Nope.

on the blithe shutting down of the machine as I have to break and run to get to work . . . and possibly dup/dupa could have run some packages and then later on in the session found some more and would start running them w/o announcing that to me the “pilot” . . . but I would need to shut the machine down? Would that interruption of dupa process cause potentially problematic outcomes? Or, no, even if I shut down while dupa is running, it will just re-start on booting TW?

Interrupting and resuming download works always. Interrupting install and reboot isn’t a good idea. You may try suspend and resume instead.

From what I saw in journalctl it looks like “dupa” shows both the dup download and then runs the remove/install of the packages . . . which didn’t exactly fly by . . . .

If cache is up to date dupa will install only. If out of date it will readily download pending packages.

For now I have two or three more TW systems to install the service, but in TW I have “stopped” it for now . . . since I come and go in and out of the system in a few hours.

Download once and upgrade many. You may keep downloaded packages on one machine and use them when upgrading the other machines.

125

I started doing this about a month ago, by bind mounting the cache on my NFS LAN server to /var/cache/zypp/packages/, which I do manually, only on boots to include duping, so only one PC at a time regardless how many TWs are running on the LAN.

126

Tried “yast2 nfs_server” on host erlangen:

**erlangen:~ #** cat /etc/exports 
/var/cache/zypp notebook(rw,root_squash,sync,no_subtree_check) 
/etc/zypp/repos.d       notebook(rw,root_squash,sync,no_subtree_check) 
**erlangen:~ #**

On host notebook I used “yast2 nfs” to mount both folders and reuse erlangen’s list of repos and zypp cache. Successfully duped the notebook from Tumbleweed 20220604-0 -> 20220606-0.

127

@karlmistelberger:

I upgraded a budget notebook (440 €) this morning:

Operating System: openSUSE Tumbleweed 20220604
KDE Plasma Version: 5.24.5
KDE Frameworks Version: 5.94.0
Qt Version: 5.15.2
Kernel Version: 5.18.1-1-default (64-bit)
Graphics Platform: X11
Processors: 8 × Intel® Core™ i5-8250U CPU @ 1.60GHz
Memory: 7.7 GiB of RAM
Graphics Processor: Mesa Intel® UHD Graphics 620

Tumbleweed 20220414-0 -> 20220604-0, 2442 packages upgraded, 40 new,  28 deleted, total size download: 1,91 GiB. Consumed 8min 46.395s CPU  time. Install consumed 13min 57.110s CPU time.

All of that data from #122 wasn’t posted at the end of my dupa.service’s maiden voyage . . . possibly just the number of packages upgraded and the CPU time . . . so essentially I posted “all of it” . . . . I tried to scroll through the posts to find out what command you used to show the full data about OS & DE but I got lost on page 2 . . . out of the 13 . . . . It’s been a process to arrive here; but thanks for your and everyone’s perseverance here.

I am now better prepared to interact with these very large recompiles . . . the next time they wash over the system . . . .

128

@karlmistelberger:

So, today is the official “TW” day . . . and the system updater applet showed “2091 packages available for upgrading” . . . . So I started the dup.service:

#systemctl start dup.service

There was a tiny delay and the cursor returned . . . . So looking at posted data I ran:

#systemctl status dup

And I got some activity, but then it continued to show the same 10 packages as downloading, over and over but saying

lines 1- 24/24 END

with the last part of END flashing like a cursor, and showing:

● dup.service - Dist Upgrade
     Loaded: loaded (/etc/systemd/system/dup.service; static)
     Active: active (running) since Mon 2022-06-20 07:24:03 PDT; 44s ago
    Process: 6601 ExecStartPre=/usr/bin/sleep 11 (code=exited, status=0/SUCCESS)
   Main PID: 6628 (Zypp-main)
      Tasks: 8 (limit: 4915)
        CPU: 8.117s
     CGroup: /system.slice/dup.service
             ├─ 6628 /usr/bin/zypper dist-upgrade --no-confirm --download-only
             ├─ 6648 gpg-agent --homedir /var/tmp/zypp.kBWjXe/zypp-trusted-krsFiq9E --use-standard-socket --daemon
             ├─ 6650 scdaemon --multi-server --homedir /var/tmp/zypp.kBWjXe/zypp-trusted-krsFiq9E
             ├─ 6674 /usr/bin/systemd-inhibit --what=sleep:shutdown:idle --who=zypp --mode=block "--why=Zypp commi>
             └─ 6675 /usr/bin/cat

Jun 20 07:24:45 linux-f6nl zypper[6628]: Retrieving: mateweather-common-1.26.0-1.8.noarch.rpm ...done (5.2 MiB/s)]
Jun 20 07:24:45 linux-f6nl zypper[6628]: Retrieving package mathjax-ams-fonts-2.6.0-3.5.noarch (51/2162), 127.9 Ki>
Jun 20 07:24:46 linux-f6nl zypper[6628]: Retrieving: mathjax-ams-fonts-2.6.0-3.5.noarch.rpm .done]
Jun 20 07:24:46 linux-f6nl zypper[6628]: Retrieving package mathjax-caligraphic-fonts-2.6.0-3.5.noarch (52/2162), >
Jun 20 07:24:46 linux-f6nl zypper[6628]: Retrieving: mathjax-caligraphic-fonts-2.6.0-3.5.noarch.rpm [done]
Jun 20 07:24:46 linux-f6nl zypper[6628]: Retrieving package mathjax-fraktur-fonts-2.6.0-3.5.noarch (53/2162), 132.>
Jun 20 07:24:46 linux-f6nl zypper[6628]: Retrieving: mathjax-fraktur-fonts-2.6.0-3.5.noarch.rpm [done]
Jun 20 07:24:46 linux-f6nl zypper[6628]: Retrieving package mathjax-main-fonts-2.6.0-3.5.noarch (54/2162), 264.2 K>
Jun 20 07:24:47 linux-f6nl zypper[6628]: Retrieving: mathjax-main-fonts-2.6.0-3.5.noarch.rpm .done]
Jun 20 07:24:47 linux-f6nl zypper[6628]: Retrieving package mathjax-math-fonts-2.6.0-3.5.noarch (55/2162), 121.7 K

I don’t know if this is still downloading, 50 minutes later, or I have to start dup.service to install . . . I’ve been waiting for the # cursor to come back, but it seems like we have jumped into a “new” terminal window that is on endless loop??

OR, the “status dup” command is running the download and install???

129

Opened a new tab and ran the start dupa . . . still showing a “synopsis” of ten packages installing from cache??

# journalctl -b -u dup.service -g download
Jun 20 07:24:06 linux zypper[6628]: Overall download size: 1.61 GiB. Already cached: 0 B. Download only.

linux:/home/butoh-mind # systemctl status dupa.service
○ dupa.service - Dist Upgrade, apply download
     Loaded: loaded (/etc/systemd/system/dupa.service; static)
     Active: inactive (dead)

linux# systemctl start dupa.service

linux # systemctl status dupa.service
● dupa.service - Dist Upgrade, apply download
     Loaded: loaded (/etc/systemd/system/dupa.service; static)
     Active: active (running) since Mon 2022-06-20 08:31:01 PDT; 22s ago
   Main PID: 16159 (Zypp-main)
      Tasks: 10 (limit: 4915)
        CPU: 22.817s
     CGroup: /system.slice/dupa.service
             ├─ 16159 /usr/bin/zypper dist-upgrade --no-confirm
             ├─ 16179 gpg-agent --homedir /var/tmp/zypp.s7pK0J/zypp-trusted-kr23TM9y --use-standard-socket --daemon
             ├─ 16181 scdaemon --multi-server --homedir /var/tmp/zypp.s7pK0J/zypp-trusted-kr23TM9y
             ├─ 16263 /usr/bin/systemd-inhibit --what=sleep:shutdown:idle --who=zypp --mode=block "--why=Zypp comm>
             ├─ 16264 /usr/bin/cat
             ├─ 16265 /bin/bash /usr/lib/zypp/plugins/commit/btrfs-defrag-plugin.sh
             └─ 16275 /usr/lib/zypp/plugins/commit/snapper-zypp-plugin

Jun 20 08:31:24 linux zypper[16159]: In cache libqt5-qttranslations-5.15.2+kde22-1.5.x86_64.rpm (31/2163),   >
Jun 20 08:31:24 linux zypper[16159]: In cache libsemanage-conf-3.3-2.7.x86_64.rpm (32/2163),  11.3 KiB (  2.3>
Jun 20 08:31:24 linux zypper[16159]: In cache libssh-config-0.9.6-3.4.x86_64.rpm (33/2163),  13.0 KiB (  159 >
Jun 20 08:31:24 linux zypper[16159]: In cache libstorage-ng-lang-4.5.21-1.1.noarch.rpm (34/2163), 190.2 KiB (>
Jun 20 08:31:24 linux zypper[16159]: In cache libthai-data-0.1.29-1.3.x86_64.rpm (35/2163), 188.3 KiB (574.3 >
Jun 20 08:31:24 linux zypper[16159]: In cache libtirpc-netconfig-1.3.2-2.9.x86_64.rpm (36/2163),  13.7 KiB ( >
Jun 20 08:31:24 linux zypper[16159]: In cache lifecycle-data-openSUSE-1-5.12.noarch.rpm (37/2163),   8.4 KiB >
Jun 20 08:31:24 linux zypper[16159]: In cache lua54-doc-5.4.4-3.2.noarch.rpm (38/2163), 111.6 KiB (420.5 KiB >
Jun 20 08:31:24 linux zypper[16159]: In cache man-pages-5.13-1.5.noarch.rpm (39/2163),   3.6 MiB (  4.5 MiB u>
Jun 20 08:31:24 linux zypper[16159]: In cache man-pages-posix-2017a-1.7.noarch.rpm (40/2163),   2.7 MiB (  2.

Doesn’t seem to be any “live” tracking of the progress?? It just keeps showing the “1 - 25 lines” with flashing “END” . . . . ???

130

The above is puzzling. Show all by running ‘journalctl -b -u dup | susepaste -t “Tumbleweed dup journal”’: SUSE Paste

131

It seems like the “status” operations were “running” but not showing any additional info . . . added a third tab after another hour or so . . . .

# journalctl -b -u dupa.service -g Consumed
Jun 20 09:28:31 linux-f6nl systemd[1]: dupa.service: Consumed 19min 18.845s CPU time.

linux # zypper clean --all
All repositories have been cleaned up.

linux # systemctl stop dup.service
linux # systemctl stop dupa.service

zypper shows “system up to date” . . . sort of “painless” but sort of “blind”???

132

@karlmistelberger:

Thanks for the reply. It does seemed to have finally completed the tasks, zypper shows “nothing to do.” But. yes, puzzling indeed . . . so I guess you EU guys use the reverse quotations of what we do . . . . Do I run

'journalctl -b -u dup | susepaste -t "Tumbleweed dup journal"'

And that will make file on my desktop, that I then paste into the linked site?? Or do I run the entire

'journalctl -b -u dup | susepaste -t "Tumbleweed dup journal"': https://paste.opensuse.org/85308372

But with the quotes only around “Tumbleweed dup journal”?? There are some issues with the single and double quotes that don’t seem to be correct??

133

Make sure you have proper susepaste defaults, see “man susepaste”. Copy the entire line (triple click to select the entire line) and paste into terminal window:

journalctl -b -u dupa | susepaste -t "Tumbleweed dup journal"

Post the displayed link here.

134

@karlmistelberger:

Thanks for the clarification . . . I may have screwed up on the “flow” . . . . I was trying to send an email via Gmail for half an hour, something in the upgrade must have broken the internet for Gmail . . . I was making my posts here in the thread, but Gmail wasn’t sending my long email out . . . . I tried this and that and then in frustration I logged out . . . and then , thinking like I might have already lost the dupa data trail . . . I shut down. :expressionless:

Did some actual tasks . . . then on cold boot, back into TW . . . now, after the 2091 package upgrade–no menu button in the toolbar . . . ??? This happened before in MATE . . . so I added the “classic menu applet” to get back into Firefox.

Will the “dupa | susepaste” data still be there or has it gone the way of the Dodo . . . and I’ll have to wait another week or so before there is another 2K package upgrade to check it??

135

@karlmistelberger:

I ran the command, but the cursor came back so quickly I don’t think there is anything much to see considering . . . . Have to catch it next time??

So, if I just want to download AND install the packages, do I just “start dupa.service” and then run the “dupa.service status” command?? It usually doesn’t take that long to download the packages, it’s the installation that seems to take zypper so much time to execute . . . .

https://susepaste.org/73074514
   https://paste.opensuse.org/73074514

Yeah . . . skip it, it said “no entries” . . . the scourge of self-sabotage strikes again . . . .

136

Paste into a root shell.

137

Urp . . . very embarassing . . . busy . . . too busy to think. This one might have some more data???

https://susepaste.org/99032095
   https://paste.opensuse.org/99032095

Still says, “No entries” even though pasted as “root” . . . . I shut the machine down, wouldn’t that wipe the journal??

138

I have no idea what you are doing. I only made a guess in post #136. Always check what you are doing. Run the first part of the command first and view its output in your terminal window. If the command prints what you think you should post, pipe it into susepaste. Post as follows, including the very command you ran:

**6700K:~ #** journalctl -b -u dupa | susepaste -t "Tumbleweed dup journal" -e 40320  
Pasted as: 
   https://susepaste.org/46644078 
   https://paste.opensuse.org/46644078 
Link is also in your clipboard. 
**6700K:~ #**
139

Hi.
About 2800 packages, python310 is already there, but I still have a python folder which is a link to a python2.7 folder.
If I try to uninstall python, the next packages will be removed:



  gimp-plugins-python python python-gobject2 python-gtk

package python-base is python-base-2.7.18-22.2.src

environment:


Kernel: 5.18.4-1-default arch: x86_64 bits: 64
    Desktop: Xfce v: 4.16.0 Distro: openSUSE Tumbleweed 20220619

How can I get rid of python 2.7, in the case is not necessary anymore, without breaking anything, or this will be fixed by the distro automatically in the near future?

140

Presumably you don’t need this. Run “zypper rm --clean-deps python”, hold your breath and type “y” at the prompt. Reinstall if fatal.