+ Reply to Thread
Page 1 of 2 12 LastLast
Results 1 to 10 of 16
  1. #1
    Join Date
    Oct 2002
    Location
    Nanaimo, BC
    Posts
    12,204
    Rep Power
    10

    Default 4.4.5.7-7834 Feedback. Post results here

    This thread was started for reports and discussion about the new 4.4.5.7-7834 firmware release. Available at: https://ln.sync.com/dl/347c254c0#2fw...dpkg7-m6ps4vta

    You should no longer, or very rarely, see the F9 logs reporting that there was too much noise or a PCI bus problem and the card is being reset. This did keep your system running and was a short/long term workaround. It is fixed and now I can concentrate on the other issues which I will list here, in order of importance.

    1. I will track down: kernel: wifi: phy0: Failed to stop TX DMA, times = 2, queues=0x003! As near as I can tell it does no real damage but I consider it to be a coding error and finding the cause might even fix some other issues.

    2. I see some ping loss and the first one two pings are typically always missed on an idle system. On an active system there is no ping miss for the first 2.
    ..a computer is a stupid machine with the ability to do incredibly smart things, while computer programmers are smart people with the ability to do incredibly stupid things. They are,in short, a perfect match..

    Try the latest 4.4.5.7-7842 release. It fixes the 11N card reset issue.
    http://www.star-os.com/ http://www.star-os.com/store

  2. #2
    Join Date
    Oct 2002
    Location
    Nanaimo, BC
    Posts
    12,204
    Rep Power
    10

    Default

    ninedd has updated one whole system with an AP and 17 clients. The really bad client is still showing MANY, MANY resets so I am not finished with it just yet.

    I went though all of the clients, and obviously the 11G units show no issues. The 11N is the only card affected. He has 9 11G and 8 11N. Of those 8 there is only 1 showing really bad numbers with about 70 resets over 3 hours. The others are all 4 or less.

    Note that it is advisable to try this if you have resets and 11N cards. The specific problem I fixed might help you. If it does not then it is easy to downgrade and wait for the next fix, BUT I would appreciate if you will test this an report your results (before and after numbers over a 4 or 5 hour period unless you have to downgrade quickly) to this thread.
    ..a computer is a stupid machine with the ability to do incredibly smart things, while computer programmers are smart people with the ability to do incredibly stupid things. They are,in short, a perfect match..

    Try the latest 4.4.5.7-7842 release. It fixes the 11N card reset issue.
    http://www.star-os.com/ http://www.star-os.com/store

  3. #3
    Join Date
    Aug 2005
    Posts
    99
    Rep Power
    13

    Default

    Tried to upgrade an APU from 4.4.5.5 to 4.4.5.7 and got no connection after upgrade. Can try again later, but I noticed size of my downloaded .pkg is substantially smaller for 5.7 than 5.5 - is that correct?

  4. #4
    Join Date
    Oct 2002
    Location
    Nanaimo, BC
    Posts
    12,204
    Rep Power
    10

    Default

    The APU is one I am unable to test at the moment. The build must be broken so I will remove it from the download until I have fixed it. I'm surprised since the ALix built with no issues and the APU is quite similar as to the hardware. Thanks for testing. I hope you were able to downgrade it.

    Quote Originally Posted by timotei View Post
    Tried to upgrade an APU from 4.4.5.5 to 4.4.5.7 and got no connection after upgrade. Can try again later, but I noticed size of my downloaded .pkg is substantially smaller for 5.7 than 5.5 - is that correct?
    ..a computer is a stupid machine with the ability to do incredibly smart things, while computer programmers are smart people with the ability to do incredibly stupid things. They are,in short, a perfect match..

    Try the latest 4.4.5.7-7842 release. It fixes the 11N card reset issue.
    http://www.star-os.com/ http://www.star-os.com/store

  5. #5
    Join Date
    Dec 2002
    Location
    Frozen Tundra of Canada
    Posts
    2,818
    Rep Power
    17

    Default

    Quote Originally Posted by lonnie View Post
    He has 9 11G and 8 11N. Of those 8 there is only 1 showing really bad numbers with about 70 resets over 3 hours. The others are all 4 or less.
    Lonnie - sometime after I upgraded everyone on that AP to 4.4.5.7-7834 that one particularly bad client (the x.x.x.136 client) who had 70 resets in ~ 3 hours - it then reset by itself (or maybe she power cycled it?) and since then, there have been 0 resets in 1 day, 13 hours.

    So, I'm not certain what to think. I don't think she's gone 1.5 days with no resets ever since the reset counter was added - so this is probably good news. Of course, we can't really graph the resets in CACTI or PRTG or anything, so I can't be 100% certain about that. But as far as I remember, she's never gone a day without any resets. However, maybe she's also made a change to her router and if so, that's screwed up our test environment too.

    Last ~ 1.5 to 2 days (depending on which customer) since upgrading everyone to 4.4.5.7-7834

    .1 = 5 resets (AP)
    .25 = 4 resets (Test in shed)
    .100 = 0 resets
    .116 = 1 reset
    .129 = 1 reset
    .136 = 0 reset
    .138 = 0 reset

    So, these numbers are way better - in the previous 2 days before upgrading to build 7834, most of the WAR1B clients had a reset count of at least 7 or more, and x.x.x.136 had 174 resets in the previous couple days. We then upgraded her - she had ~ 70 resets in about 3 hours, then it spontaneously rebooted (or she power cycled) and since then she's had 0 resets.
    Last edited by ninedd; 04-13-2016 at 11:18 PM.

  6. #6
    Join Date
    Oct 2002
    Location
    Nanaimo, BC
    Posts
    12,204
    Rep Power
    10

    Default

    I've made good progress but the job is not quite done. This issue has been bugging the Linux guys since 2013 with very little success.

    The frustrating part is that the symptoms come and go as you pointed out regarding the .136 client.

    I am creating massive noise which was giving me resets but yesterday I saw 0 resets, even when dropping back to 7825 which basically has no fixes. This morning I was able to see about 4 per hour so I have made a few more changes, and no resets in 30 minutes. The question is did my patch fix it or did the symptom go away for a bit? I'll watch it for a bit and then downgrade to see if the older release gives resets.

    At the end of the day I believe that 7834 is better than what is out there now, so I advise people to apply it. You can always downgrade if you feel that it makes things worse, but it might just improve things. It is not yet perfect but it is moving in the right direction.

    Quote Originally Posted by ninedd View Post
    Lonnie - sometime after I upgraded everyone on that AP to 4.4.5.7-7834 that one particularly bad client (the x.x.x.136 client) who had 70 resets in ~ 3 hours - it reset by itself (or maybe she power cycled it?) and since then, there have been 0 resets in 1 day, 13 hours.

    So, I'm not certain what to think. I don't think she's gone 1.5 days with no resets ever since the reset counter was added - so this is probably good news. Of course, we can't really graph the resets in CACTI or PRTG or anything, so I can't be 100% certain about that. But as far as I remember, she's never gone a day without any resets. However, maybe she's also made a change to her router and if so, that's screwed up our test environment too.

    Last ~ 1.5 to 2 days (depending on which customer) since upgrading everyone to 4.4.5.7-7834

    .1 = 5 resets (AP)
    .25 = 4 resets (Test in shed)
    .100 = 0 resets
    .116 = 1 reset
    .129 = 1 reset
    .136 = 0 reset
    .138 = 0 reset

    So, these numbers are way better - in the previous 2 days before upgrading to build 7834, most of the WAR1B clients had a reset count of at least 7 or more, and x.x.x.136 had 174 resets in the previous couple days. We then upgraded her - she had ~ 70 resets in about 3 hours, then it spontaneously rebooted (or she power cycled) and since then she's had 0 resets.
    ..a computer is a stupid machine with the ability to do incredibly smart things, while computer programmers are smart people with the ability to do incredibly stupid things. They are,in short, a perfect match..

    Try the latest 4.4.5.7-7842 release. It fixes the 11N card reset issue.
    http://www.star-os.com/ http://www.star-os.com/store

  7. #7
    Join Date
    Dec 2002
    Location
    Frozen Tundra of Canada
    Posts
    2,818
    Rep Power
    17

    Default

    Our AP #86 went to 1 association. The only CPE associated was the 'test in shed', but the rest all dissociated - and the one CPE still associated was unreachable. I saved support and sysinfo and kernlog files. I'll email them to you.

  8. #8
    Join Date
    Oct 2002
    Location
    Nanaimo, BC
    Posts
    12,204
    Rep Power
    10

    Default

    Thanks I am going through them now.
    ..a computer is a stupid machine with the ability to do incredibly smart things, while computer programmers are smart people with the ability to do incredibly stupid things. They are,in short, a perfect match..

    Try the latest 4.4.5.7-7842 release. It fixes the 11N card reset issue.
    http://www.star-os.com/ http://www.star-os.com/store

  9. #9
    Join Date
    Dec 2002
    Location
    Frozen Tundra of Canada
    Posts
    2,818
    Rep Power
    17

    Default

    I don't remember if I told you or not - I've set that AP#86 into '11g' mode instead of 'HT20' mode, just to test that over the weekend. That's gone 1x1 of course on AP side. I didn't make any changes to the clients, but I just thought that maybe if the AP ran differently or better or worse in '11g' mode, then maybe that would give another theory or idea. I'll switch it back to 'HT20 2x2' on Monday and see over a few days if that changes it's behavior, and maybe add another clue in the clue pile.

  10. #10
    Join Date
    Oct 2002
    Location
    Nanaimo, BC
    Posts
    12,204
    Rep Power
    10

    Default

    I did see that. I note that it still gets resets. It will be interesting to see if it stalls and needs an activate.
    ..a computer is a stupid machine with the ability to do incredibly smart things, while computer programmers are smart people with the ability to do incredibly stupid things. They are,in short, a perfect match..

    Try the latest 4.4.5.7-7842 release. It fixes the 11N card reset issue.
    http://www.star-os.com/ http://www.star-os.com/store

Similar Threads

  1. 4.4.5.7-7834 has been released
    By lonnie in forum StarOS™ BETA Releases
    Replies: 0
    Last Post: 04-11-2016, 10:24 PM
  2. Older hardware and firmware feedback on versions tried and results.
    By David L. Vrablic in forum General Discussion
    Replies: 29
    Last Post: 03-03-2011, 01:56 AM
  3. Good service feedback
    By DLNoah in forum General Discussion
    Replies: 3
    Last Post: 08-21-2008, 12:25 AM
  4. Feedback
    By Steve in forum Steve Dietzel's StarCom & UtiliStar
    Replies: 1
    Last Post: 05-23-2004, 04:30 AM
  5. Squid Feedback!
    By Steve in forum Feature Requests
    Replies: 2
    Last Post: 11-07-2002, 05:19 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts