Groups keyboard shortcuts have been updated
Dismiss
See shortcuts

SWUFORWARD - swupdate doesn't return on subsystem - parent gets stuck

78 views
Skip to first unread message

Armen

unread,
Jul 19, 2024, 10:32:30 AM7/19/24
to swupdate
I have 4 subsystems that I am trying to update using SWUFORWARD (with swupdate 2019.11).
I am facing a problem where subsystem never returns after swupdate is completed. Parent system gets stuck waiting and doesn't go to the next subsystem (I have 4 image sections for each subsystem to do update 1 by 1).

swuforward has no problem connecting or sending the file. At least one system always gets updated. It can get stuck either on 1st or 2nd subsystem - this is random. But main problem is that last subsystem never return done. It will show "SUCCESS !" from swupdate-progress but never update parent swupdate process that it is completed.

Is it a know problem and how do I resolve it?

Armen

unread,
Jul 19, 2024, 11:24:15 AM7/19/24
to swupdate
A successful swuforward to subsystem will display trace indicating start, run, success.

[TRACE] : SWUPDATE running :  [callback_ws_swupdate] : Connection to http://10.0.0.5:8080: established
< HTTP/1.1 200 OK
< Server: Mongoose/6.11
< Content-Type: text/plain
< Connection: close
<
Ok, swupdate.swu - 271152128 bytes.
* Closing connection 0
[TRACE] : SWUPDATE running :  [callback_ws_swupdate] : Change status on http://10.0.0.5:8080 : START
[TRACE] : SWUPDATE running :  [callback_ws_swupdate] : Change status on http://10.0.0.5:8080 : RUN
[TRACE] : SWUPDATE running :  [callback_ws_swupdate] : Change status on http://10.0.0.5:8080 : SUCCESS
[TRACE] : SWUPDATE running :  [install_single_image] : Found installer for stream check_id.jbc.enc cpld_id

Unsuccessful swuforward to subsystem will only display "Closing connection 0". No other updates.
< HTTP/1.1 200 OK
< Server: Mongoose/6.11
< Content-Type: text/plain
< Connection: close
<
Ok, swupdate.swu - 271152128 bytes.
* Closing connection 0

After this, parent swupdate process is stuck in waiting.

Stefano Babic

unread,
Jul 21, 2024, 9:51:29 AM7/21/24
to Armen, swupdate
Hi Armen,

first at all there is no support for older release, you have a 5 years
old version. Developmen can be done on the current release.
Mongoose was update several times in between. The log just says that the
Websocket communication to some underlying devices is not working. The
master (device running the swuforward handler) then waits for packets.
It is expected that, if the connection could be established, devices
will send the result. Here it seems it is not the case, I cannot say the
reason, and no, this does not happen. It should be analyzed on your
environment what is really happening.

Best regards,
Stefano Babic

> After this, parent swupdate process is stuck in waiting.
> On Friday, July 19, 2024 at 9:32:30 AM UTC-5 Armen wrote:
>
> I have 4 subsystems that I am trying to update using SWUFORWARD
> (with swupdate 2019.11).
> I am facing a problem where subsystem never returns after swupdate
> is completed. Parent system gets stuck waiting and doesn't go to the
> next subsystem (I have 4 image sections for each subsystem to do
> update 1 by 1).
>
> swuforward has no problem connecting or sending the file. At least
> one system always gets updated. It can get stuck either on 1st or
> 2nd subsystem - this is random. But main problem is that last
> subsystem never return done. It will show "SUCCESS !" from
> swupdate-progress but never update parent swupdate process that it
> is completed.
>
> Is it a know problem and how do I resolve it?
>
> --
> You received this message because you are subscribed to the Google
> Groups "swupdate" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to swupdate+u...@googlegroups.com
> <mailto:swupdate+u...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/swupdate/dea6381c-701d-437b-a3fd-6d34e02d7a6an%40googlegroups.com <https://groups.google.com/d/msgid/swupdate/dea6381c-701d-437b-a3fd-6d34e02d7a6an%40googlegroups.com?utm_medium=email&utm_source=footer>.

Armen

unread,
Jul 29, 2024, 3:31:01 PM7/29/24
to swupdate
Hi Stefano,
If main swupdate process hangs on image step that fails to get DONE reply from web server on subsystem, how do I go about terminating that step and continuing with main swupdate process?
Or is it possible to have timeout for any image block in swupdate process so that error like that will not able to hang main swupdate process indefinitely?

Stefano Babic

unread,
Jul 30, 2024, 9:28:50 AM7/30/24
to Armen, swupdate
On 29.07.24 21:31, Armen wrote:
> Hi Stefano,
> If main swupdate process hangs on image step that fails to get DONE
> reply from web server on subsystem, how do I go about terminating that
> step and continuing with main swupdate process?
> Or is it possible to have timeout for any image block in swupdate
> process so that error like that will not able to hang main swupdate
> process indefinitely?

As I said, this should be already detected. On the other side there is a
SWUpdate Webserver, and if the connection is working, it is pushing data
to the Websocket connection. And if connection gets broken, it should be
detected the case as well. Why it is not working on your side, it is
topic for debugging.

Best regards,
Stefano Babic

>
> On Sunday, July 21, 2024 at 8:51:29 AM UTC-5 Stefano Babic wrote:
>
> Hi Armen,
>
> first at all there is no support for older release, you have a 5 years
> old version. Developmen can be done on the current release.
>
> On 19.07.24 17:24, Armen wrote:
> > A successful swuforward to subsystem will display trace indicating
> > start, run, success.
> >
> > [TRACE] : SWUPDATE running :  [callback_ws_swupdate] : Connection to
> > http://10.0.0.5:8080 <http://10.0.0.5:8080>: established
> > < HTTP/1.1 200 OK
> > < Server: Mongoose/6.11
> > < Content-Type: text/plain
> > < Connection: close
> > <
> > Ok, swupdate.swu - 271152128 bytes.
> > * Closing connection 0
> > [TRACE] : SWUPDATE running :  [callback_ws_swupdate] : Change
> status on
> > http://10.0.0.5:8080 <http://10.0.0.5:8080> : START
> > [TRACE] : SWUPDATE running :  [callback_ws_swupdate] : Change
> status on
> > http://10.0.0.5:8080 <http://10.0.0.5:8080> : RUN
> > [TRACE] : SWUPDATE running :  [callback_ws_swupdate] : Change
> status on
> > http://10.0.0.5:8080 <http://10.0.0.5:8080> : SUCCESS
> https://groups.google.com/d/msgid/swupdate/dea6381c-701d-437b-a3fd-6d34e02d7a6an%40googlegroups.com <https://groups.google.com/d/msgid/swupdate/dea6381c-701d-437b-a3fd-6d34e02d7a6an%40googlegroups.com> <https://groups.google.com/d/msgid/swupdate/dea6381c-701d-437b-a3fd-6d34e02d7a6an%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/swupdate/dea6381c-701d-437b-a3fd-6d34e02d7a6an%40googlegroups.com?utm_medium=email&utm_source=footer>>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "swupdate" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to swupdate+u...@googlegroups.com
> <mailto:swupdate+u...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/swupdate/81b6ab17-546c-419e-8829-fc1ae40564afn%40googlegroups.com <https://groups.google.com/d/msgid/swupdate/81b6ab17-546c-419e-8829-fc1ae40564afn%40googlegroups.com?utm_medium=email&utm_source=footer>.

Armen

unread,
Aug 1, 2024, 5:05:19 PM8/1/24
to swupdate
Connection seems to get lost after the file is send to submodule. This happens often (~ maybe 80% of times).

So submodule continues with performing it's swupdate, but main system never gets updates from it. Main system ends up waiting on submodule to get back but this never happens.

When connection is not lost, main system will receive all updates from submodule during the update.


Reply all
Reply to author
Forward
0 new messages