The KiwiSDR 2 online store is open for orders! Please visit kiwisdr.nz

v1.707

jksjks
edited November 3 in Software release information

UPDATE: For now please do not follow the instructions below. The v1.707 version I just released seems to be updating okay. It is not using the binary update feature so will take 20 - 30 minutes to build.

----

Unfortunately, the github.com problems we've had with Kiwis using the original Debian 8 distro have now spread to all versions of Debian (i.e. 9, 10, 11).

A fix for this issue is in the v1.706 release. But the problem is your Kiwi running a version prior to v1.706 can't auto-update to get the fix. You have to manually force an update to get v1.706. After that auto-updates can continue (hopefully).

Proceed as follows:

  • Go to the admin update tab. Click the "Check now" button. If the status message says "Git clone damaged!" then click the button again. You should see a spinning icon and "Checking for software update". It may take a full minute or two for the spinning to stop and the message "Available version: v1.706" (or later) to appear. It is possible you may have to click the button several times if you continue to get "Git clone damaged!" messages. Some have reported needing to click the button almost a dozen times.
  • Now click the "Build now" button . The Kiwi will restart when the build completes in 20 - 30 minutes.

When the Kiwi restarts it should be running v1.707 (or later).

«1

Comments

  • ArturPLArturPL
    edited November 2

    Kiwi V2 Debian 11.11the message about the available version appeared after 4 clicks, but now I see this:

    kiwi.JPG

    I'm trying now with Kiwi V1 D11.11 but I'm clicking 9 times and still Git clone damaged!

    -edit: Kiwi v1 worked after 11 clicks and started updating, Kiwi v2 still the same: Filesystem is FULL!

    edit: 2 Kiwi v1 Installed 1.706 in about 5 minutes and works stably

    edit3: the last one reacted after !27 or 28! clicks, I've already lost count

  • jksjks
    edited November 2

    For the Kiwi-2, it's possible that the log files are huge. Possibly from update retries and also a bot that has been connecting repeatedly on a number of Kiwis filling up the logs.

    In the admin console type "m clean_logs" and note the amount of before and after filesystem free space. Also type "d." to see the actual filesystem free space.

    For the Kiwi-1 try in the admin console the command "git fetch origin". If that shows error messages try again a few times. Paste errors here or email to support@kiwisdr.com

  • ArturPLArturPL 
    root@kiwisdr:~/Beagle_SDR_GPS# m clean_logs
journalctl --disk-usage
Archived and active journals take up 352.0M in the file system.
journalctl --vacuum-size=10M
Vacuuming done, freed 0B of archived journals from /var/log/journal.
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000ddc-000625f054d245db.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000de2-000625f05bd60dd0.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000de3-000625f05bd608e2.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000de5-000625f05bd60e89.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000de7-000625f05bd607e6.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000de9-000625f05bd60d9f.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000df1-000625f062b9c785.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000df4-000625f062b9c2cf.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000df6-000625f062b9c78f.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000dfc-000625f06d1f57de.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000dff-000625f06d1f5777.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000e00-000625f06d1f5723.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000e02-000625f06d1f56f3.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000e07-000625f074b9d130.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000e09-000625f074b9d928.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000e0b-000625f074b9d8d3.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000e0d-000625f074b9d301.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000e14-000625f07b94f804.journal (8.0M).
Deleted archived journal /var/log/journal/5436a1d97ac646fe91d742fc5430cb34/system@1bb179a086f14d0aaa5396b6c3ab996d-0000000000000e15-000625f07b94f32e.journal (8.0M).
Vacuuming done, freed 336.0M of archived journals from /var/log/journal/5436a1d97ac646fe91d742fc5430cb34.
Vacuuming done, freed 0B of archived journals from /run/log/journal.
journalctl --disk-usage
Archived and active journals take up 16.0M in the file system.
root@kiwisdr:~/Beagle_SDR_GPS#  


  • ArturPLArturPL 
    root@kiwisdr:~/Beagle_SDR_GPS# df -h .; journalctl --vacuum-size=10M; df -h .
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p1  3.5G  2.9G  350M  90% /
Vacuuming done, freed 0B of archived journals from /run/log/journal.
Vacuuming done, freed 0B of archived journals from /var/log/journal.
Vacuuming done, freed 0B of archived journals from /var/log/journal/5436a1d97ac646fe91d742fc5430cb34.
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p1  3.5G  2.9G  350M  90% /
root@kiwisdr:~/Beagle_SDR_GPS#  

    clean logs in console solved the problem, Kiwi V2 started updating

  • jksjks

    Thanks for the report.

    11 clicks for Kiwi-1 D11.11? That's not good. Means my workaround in v1.706 may not be good enough (I just retry 3 times which is what I observed to work).

    I should probably do the log clean automatically the first time a full filesystem is detected.

  • SM7IUNSM7IUN

    I had to click six times to get a clean clone.

  • jksjks

    We're just going to have to move off of github.com onto our own hosted Git server. But that's a large amount of work and comes at a very, very bad time. I probably should have done that much sooner..

  • jksjks
    edited November 2

    Okay, let me do a v1.707 release that addresses both these issues before too many people follow this procedure. Thank you!

  • jksjks
    edited November 3

    Ugh. This situation is much worse than I thought.

    I forgot that git keeps deltas of all changes to files so you can step to an arbitrary commit date and see the files as they were. But this means for the new ~30MB binary update files it simply keeps all the old ones. It doesn't delete them. Now the total git clone size is expanding at an alarming rate. This is almost certainly a contributor to the recent problems.

    I'll need some time to fix this.

  • smgsmg

    Is it worth standing up a private git?

    I might be able to help as I have a proxmox cluster in Auckland.

  • jksjks

    I've already got two VPS that I'm paying for (website, proxy service, forum, file server, TDoA backend, ...) Plenty of storage, but the bandwidth costs are depressing (over 3.5TB/month now due to the included proxy service with Kiwi-2).

    What's scaring the crap out of me are these tools that are supposed to help remove large blobs from a git repo (git-filter-repo et al). It seems you can do great damage to your repo if you're not careful. And I don't really understand the subtleties.

    I think it might be simpler to just start over with a new repo without the large files. And change all the code that references large files to grab them off kiwisdr.com out-of-tree. Like what we do now for things like the latest IP blacklist and the various frequency lists. There are things like Git Large File Storage. But I'm not sure that has any real advantage.

    After running git-largest-files I found there are big files that were deleted long ago. Compressed versions of these are getting pulled over every time you do a git clone. No wonder the current .git directory is 358MB in size.

  • smgsmg

    are you able to rate limit any of that traffic?

  • jksjks

    No. But I get charged on bulk transfer amount, not rate. The rate is relatively low since it's just x-hundreds of Kiwi audio/waterfall channels. Averages roughly 20Mb/s (both in/out since it's mostly proxy traffic vs everything else).

  • smgsmg

    How much storage do you need?

  • HB9TMCHB9TMC

    but the bandwidth costs are depressing (over 3.5TB/month

    3.5 TB/month doesn't seem to be much.

    My cloud server has 20 TB/month included and it's just 4€ a month. Maybe it's worth looking for an alternative.

  • jksjks
    edited November 3

    Thanks guys, but my VPS service is not the priority at the moment. I have to get this disastrous git situation under control before every Kiwi on the planet becomes bricked.

    smg
  • smgsmg
    https://forum.kiwisdr.com/index.php?p=/discussion/comment/19908#Comment_19908

    If we can help in any way - reach out.

  • HolgerHolger
    edited November 6

    In my case i had to rebuild the Kiwisoftware completly via SSH as described here:

    https://forum.kiwisdr.com/index.php?p=/discussion/3184/rebuilding-the-kiwi-software

    Took hours but now i am on 1.707.


    73 Holger, DF6DBF

  • ShaneBrookShaneBrook

    I get this:

    Installed version: v1.702, built Oct 23 2024 20:33:26

    Filesystem is FULL!

    I tried this:

    CONSOLE: open connection
root@kiwisdr:~/Beagle_SDR_GPS# df -h .
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p1  3.6G  3.4G     0 100% /
root@kiwisdr:~/Beagle_SDR_GPS# m clean_logs
journalctl --disk-usage
Archived and active journals take up 3.0M in the file system.
journalctl --vacuum-size=10M
Vacuuming done, freed 0B of archived journals from /run/log/journal/c3aa033c1cf28d830b56a08964549473.
journalctl --disk-usage
Archived and active journals take up 3.0M in the file system.
root@kiwisdr:~/Beagle_SDR_GPS#  df -h .; journalctl --vacuum-size=10M; df -h 
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p1  3.6G  3.4G     0 100% /
Vacuuming done, freed 0B of archived journals from /run/log/journal/c3aa033c1cf28d830b56a08964549473.
Filesystem      Size  Used Avail Use% Mounted on
udev            214M     0  214M   0% /dev
tmpfs            49M  3.3M   46M   7% /run
/dev/mmcblk1p1  3.6G  3.4G     0 100% /
tmpfs           242M     0  242M   0% /dev/shm
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs           242M     0  242M   0% /sys/fs/cgroup
root@kiwisdr:~/Beagle_SDR_GPS#  


  • jksjks

    Well, your filesystem is full. See that "Avail 0, Use% 100%" in there? Have to figure out why that is.

    Start by doing "cd; du" and see who is hogging the space.

  • smgsmg

    Mine updated fine recently using the original method, check for update, rebuild. (actually before i read this post)

  • ShaneBrookShaneBrook
    https://forum.kiwisdr.com/index.php?p=/discussion/comment/19926#Comment_19926

    How to fix ?

  • ShaneBrookShaneBrook

    How to fix?

  • dl7awldl7awl

    First step, as John said:

    Start by doing "cd; du" and see who is hogging the space.

    Did you?

  • ShaneBrookShaneBrook

    I got it fived....

    I found a link here.... can't find now.....

    I did this:

    M gil

    mc

    m

    mi

    Then everything OK..... !!!

  • jksjks
    edited November 7

    For future reference, "M gil" is invalid. Perhaps "m git"? Although this can be done equivalently by the "gup" alias (git update).

  • ShaneBrookShaneBrook

    Look at this:

    CONSOLE: open connection
df -h .
root@kiwisdr:~/Beagle_SDR_GPS# df -h .
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p1  3.6G  3.3G   35M  99% /
root@kiwisdr:~/Beagle_SDR_GPS# df -h .; journalctl --vacuum-size=10M; df -h .
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p1  3.6G  3.3G   35M  99% /
Vacuuming done, freed 0B of archived journals from /run/log/journal/c3aa033c1cf28d830b56a08964549473.
Filesystem      Size  Used Avail Use% Mounted on
/dev/mmcblk1p1  3.6G  3.3G   35M  99% /
root@kiwisdr:~/Beagle_SDR_GPS# cd; du
4.0K   build.log
4.0K   makeuserlist.sh
35M    kiwi.config
104M   build
832M   Beagle_SDR_GPS
root@kiwisdr:~#  

    I need to free up space, How?

  • jksjks
    edited November 7

    Ugh. There's got to be some other log file(s) taking space for there to only be 35MB of free space. When I do this on a Kiwi running Debian 8.5 I have 1.4GB free with a /root looking like this.

    Try: "cd /var/log; du". Show us the entries that have numbers in the first column ending in "M" or "G".

  • ShaneBrookShaneBrook 
    root@kiwisdr:~# cd /var/log; du
0      btmp
0      btmp.1
4.0K   alternatives.log
4.0K   alternatives.log.1
4.0K   alternatives.log.2.gz
4.0K   alternatives.log.3.gz
4.0K   auth.log.2.gz
4.0K   auth.log.3.gz
4.0K   auth.log.4.gz
4.0K   debug
4.0K   debug.2.gz
4.0K   debug.3.gz
4.0K   debug.4.gz
4.0K   dpkg.log
4.0K   dpkg.log.2.gz
4.0K   dpkg.log.3.gz
4.0K   dpkg.log.4.gz
4.0K   dpkg.log.5.gz
4.0K   nginx
4.0K   private
8.0K   daemon.log.3.gz
8.0K   robotcontrol
12K    daemon.log.4.gz
12K    dpkg.log.1
12K    kern.log.3.gz
12K    syslog.6.gz
16K    daemon.log.2.gz
16K    debug.1
16K    syslog.3.gz
20K    kern.log.4.gz
20K    syslog.4.gz
24K    faillog
24K    syslog.5.gz
28K    kern.log.2.gz
28K    syslog.2.gz
32K    auth.log
60K    auth.log.1
60K    dpkg.log.6.gz
64K    syslog.7.gz
68K    user.log.4.gz
88K    messages.4.gz
96K    daemon.log
104K   syslog.1
120K   user.log.2.gz
124K   apt
140K   kern.log
144K   messages.2.gz
156K   user.log.3.gz
164K   messages.3.gz
260K   bootstrap.log
288K   lastlog
292K   syslog
652K   wtmp
712K   user.log
740K   daemon.log.1
808K   messages
976K   kern.log.1
1.3M   user.log.1
2.0M   messages.1
root@kiwisdr:/var/log#  


  • jksjks

    Well, it's not there. Who knows what you've done.

    You should probably start over. Re-flash from scratch using the latest Debian 11.9 image by following the instructions here: http://kiwisdr.com/info/#id-net-reflash

Sign In or Register to comment.