update-loop with sce-update -a

[Jason W]

Uploaded a new RC, please test.

[netnomad]

hi jason,

tc@box:~$ version -r
You are running dCore-jessie:2015.12.29.17.46 which is the latest release candidate.

100-desktop-jessie updated without loop.
200-multimedia-jessie updated with a loop.

thank you for your help.

[Jason W]

Did you re-run sce-update against 200-multimedia-jessie.  I expect the first run to report issues, and after that for it to work.  That is what it has done across 2 dCore installs on my box with various extensions and various .dep and lst files.

[nitram]

Unfortunately updating to dCore-jessie:2015.12.29.17.46 was not successful. After numerous updates system failed to reboot to graphics, needed backup restore.

Noted discrepancy between  sce-update -c  and  sce-update -a  , they do not appear to use a shared update check function.

Current SCEs:
Xprogs.sce
geany.sce
bsdgames-nonfree.sce
graphics-3.16.6-tinycore.sce
conky.sce 
iceweasel.sce
dbus-x11.sce 
qpdfview.sce
dillo.sce   
emelfm.sce   
xorg-intel.sce
emelfm.sce.md5.txt 
fluxbox.sce

sce-update -c flagged updates:
Xprogs
bsdgames-nonfree
dbus-x11
geany
qpdfview
xorg-intel

sce-update -a updates performed:
Xprogs
conky
dbus-x11
dillo
emelfm
fluxbox
graphics-3.16.6-tinycore
iceweasel
qpdfview
xorg-intel

Most of these updates should not have been required, some were pre-built and some, like conky, were just installed today.

Other notes:
1) Running  sce-update -c  , select '1' repeatedly to select item for check, list of available SCEs to check shortens, when get to last two items, selecting second last item automatically adds the last item to be checked rather than return to check/quit prompt.

2) Nice to have an  sce-update -c  'all' option in addition to just selecting individual items.

3) During  sce-update -a  would be very nice to have a (s)kip to next item option when this appears during update checks: 'dillo is about to be updated.  Press Enter to proceed, y to view package updates, or Ctrl-C to abort..'. Especially helpful when troubleshooting this update loop issue, to avoid rebuilding system every test.

4) With dillo example above, pressing y to view package updates then (q)uit automatically proceeds with the update. What's if the user viewed package updates and decided to abort? Would it be preferred to return to Ctrl-C abort? Especially if there was a (s)kip option, as abort kills the entire process.

Thanks for your hard work Jason, eventually you'll get the loop mystery sorted

[Jason W]

I see issues in my dCore-trusty setup.  Please hold on testing, and thank you for our patience.  I will fix soon.

[Jason W]

Can you do "sce-import conky" and then "sce-update conky"

I see no avilable updates for conky that way.  Just wanting to cover every angle.

[nitram]

This way it works as expected.

mv conky.sce conky.sce.orig
mv conky.sce.md5.txt conky.sce.md5.txt.orig

sce-import conky:

Code: [Select]

Select Package for conky

         1. conky
         2. conky-all
         3. conky-all-dbg
         4. conky-cli
         5. conky-cli-dbg
         6. conky-std
         7. conky-std-dbg

Enter selection ( 1 - 7 ) or (q)uit: 1
 
Importing conky.
 
 
Using Package Index: /etc/sysconfig/tcedir/import/debinx/debian_jessie_main_i386_Packages
 
Using Package Index: /etc/sysconfig/tcedir/import/debinx/debian_jessie_security_i386_Packages
 
Using debian Mirror: http://ftp.us.debian.org/debian
 
Using debian Security Mirror: http://ftp.us.debian.org/debian

conky is a standard debian jessie package.
Gathering dependency info..
Merging libxrender1
.
.
.
Merging conky
Parallel mksquashfs: Using 1 processor
Creating 4.0 filesystem on /etc/sysconfig/tcedir/sce/conky.sce, block size 4096.
[=============================================================-] 9679/9679 100%
Exportable Squashfs 4.0 filesystem, gzip compressed, data block size 4096
        compressed data, compressed metadata, compressed fragments, no xattrs
        duplicates are removed
Filesystem size 17247.75 Kbytes (16.84 Mbytes)
        47.65% of uncompressed filesystem size (36199.20 Kbytes)
Inode table size 31976 bytes (31.23 Kbytes)
        39.96% of uncompressed inode table size (80018 bytes)
Directory table size 11680 bytes (11.41 Kbytes)
        43.80% of uncompressed directory table size (26665 bytes)
Number of duplicate files found 53
Number of inodes 1292
Number of files 960
Number of fragments 106
Number of symbolic links  90
Number of device nodes 0
Number of fifo nodes 0
Number of socket nodes 0
Number of directories 242
Number of ids (unique uids + gids) 2
Number of uids 1
        root (0)
Number of gids 2
        root (0)
        staff (50)
Done.

sce-update -c:

Code: [Select]

conky
 
You are about to check for updates to the above SCEs. y to continue, n to exit. (y/N): y
debian
#################### 100.0% 29.4 kBps DONE     

verifying download...checksum matches OK
used 10647552 local, fetched 0
#################### 100.0% 0.0 kBps DONE   

verifying download...checksum matches OK
used 301056 local, fetched 0
Using the repo http://security.debian.org jessie main
 
Checking for updates for conky.sce..
.
.
.
No updates available for chosen SCEs at this time.

sce-update conky:

Code: [Select]

tc@box:~$ sce-update conky
debian
#################### 100.0% 0.0 kBps DONE   

verifying download...checksum matches OK
used 10647552 local, fetched 0
#################### 100.0% 0.0 kBps DONE   

verifying download...checksum matches OK
used 301056 local, fetched 0
Using the repo http://security.debian.org jessie main
 
Searching for available updates for conky.sce.
.
.
.
No updates available for chosen SCEs at this time.


[netnomad]

hi jason,

tc@box:~$ version -r
You are running dCore-jessie:2015.12.29.17.46 which is the latest release candidate.

the last cut broke again my xorg :-(

thank you for your help.

[Jason W]

Does re-importing with sce-import rather than sce-update fix it?  It should as sce-import is not broken when used by itself.

[Jason W]

I am going to go back to sce-import being called after sce-update completed like before.  Breaking sce-import is not cool and it is solid when used on it's own.  I did try to cut down on update time, but updating is something done in the background or while we are doing other things anyway.  I will upload a new cut that will not break anything when import is ran, and we have time to work on the update glitches.  Things being flagged for update when not needed or being overlooked when updates are available are not that big a deal, but breaking sce-import is.

[Jason W]

Uploaded new RC.  Should find updates on first run, once re-imported should find no updates. 

[Jason W]

There is still a glitch in how sce-update interacts with sce-import causing an error message.  I will fix tonight.

Sce-import on it's own should be just fine as before.

EDIT: Was a simple == instead of != when checking exit status of sce-import.  Testing the fix and hopefully this thing draws to a close very soon.

[Jason W]

Uploaded new RC, hopefully resolved.

[nitram]

dCore-jessie:2015.12.31.11.11


Current SCEs:
Xprogs.sce
beaver.sce
bsdgames-nonfree.sce
conky.sce
dillo.sce
emelfm.sce
fluxbox.sce
graphics-3.16.6-tinycore.sce
iceweasel.sce
xorg-intel.sce


Running sce-update -c, completed extremely quickly, only listed Xprogs for update:
Xprogs
The above SCEs need to be updated.


Running sce-update -a:
Xprogs
bsdgames-nonfree
xorg-intel
The above SCEs are about to be updated.


Then re-ran sce-update -c after updating, this time completed very slowly (probably actually checking):
No updates available for chosen SCEs at this time.


Rebooted to graphics !!!


Re-ran sce-update -c, this time completed very slowly:
No updates available for chosen SCEs at this time.


Re-ran sce-update -a:
No updates available.


Lookin good Jason.


Other:

Query typo line #23 of importupdatecheck script:
#ls /tmp/*.deb2sce > /dev/null 2>&1 && sudo rm /tmp/.*deb2sce
ls /tmp/*.deb2sce > /dev/null 2>&1 && sudo rm /tmp/*.deb2sce

Due to aterm update, now have three terminals (aterm, uxterm, xterm), does dCore base need all three?

Minor other issues noted in earlier post above.

[Jason W]

sce-update -c the first time was checking against a corrupt SCE directory.  The second time, against a good one.  So we can go with the results of the second run.
 
Are you saying line 23 of importupdatecheck can be commented out?

uxterm is part of the xterm package.  aterm has been added so the cpanel works as expected as in Core.