2:47 a.m.
Hello Thomas!
In my nightly testing I am encountering connection problems with
http://sync.memotoo.com/syncML - the connection can be establish, but
then gets disconnected by the server.
As far as I can tell, this has failed consistently for at least two
days. I did not run tests before that for a while. Any idea?
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
3:10 a.m.
Hello Patrick!
Strange ...
Do you use IPv6 ?
Does your IP change during the sync ?
What is your IP ?
--
See you soon on Memotoo !
Thomas Pequet - Webmaster Memotoo
https://www.memotoo.com/profile/thomas.memotoo
Le 21/11/2013 16:47, Patrick Ohly a écrit :
Hello Thomas!
In my nightly testing I am encountering connection problems with
http://sync.memotoo.com/syncML - the connection can be establish, but
then gets disconnected by the server.
As far as I can tell, this has failed consistently for at least two
days. I did not run tests before that for a while. Any idea?
3:32 a.m.
On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
Hello Patrick!
Strange ...
Do you use IPv6 ?
Could be. At least the machine and the transparent proxy it sits behind
are IPv6 enabled. I'm less sure whether libsoup uses IPv6.
Does your IP change during the sync ?
No.
What is your IP
curl -v http://whatismyv6.com/
...
You are connecting with an IPv6 Address of:
2001:19d0:3:5::100
...
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
4:14 a.m.
Le 21/11/2013 17:32, Patrick Ohly a écrit :
On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
> Hello Patrick!
>
> Strange ...
> Do you use IPv6 ?
Could be. At least the machine and the transparent proxy it sits behind
are IPv6 enabled. I'm less sure whether libsoup uses IPv6.
Ok I havejust change
the IPv6 so I think it is that.
Can you try again ?
> Does your IP change during the sync ?
No.
> What is your IP
curl -v http://whatismyv6.com/
...
You are connecting with an IPv6 Address of:
2001:19d0:3:5::100
...
6:39 a.m.
On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
Le 21/11/2013 17:32, Patrick Ohly a écrit :
> On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
> > Hello Patrick!
> >
> > Strange ...
> > Do you use IPv6 ?
> Could be. At least the machine and the transparent proxy it sits behind
> are IPv6 enabled. I'm less sure whether libsoup uses IPv6.
Ok I havejust change the IPv6 so I think it is that.
Can you try again ?
Still fails.
Note to self: a simple test is "curl http://www.memotoo.com/syncmldev"
With IPv4, I get a redirect. With IPv6, it just hangs.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
7:35 p.m.
Le 21/11/2013 20:39, Patrick Ohly a écrit :
On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
> Le 21/11/2013 17:32, Patrick Ohly a écrit :
>
>> On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
>>> Hello Patrick!
>>>
>>> Strange ...
>>> Do you use IPv6 ?
>> Could be. At least the machine and the transparent proxy it sits behind
>> are IPv6 enabled. I'm less sure whether libsoup uses IPv6.
> Ok I havejust change the IPv6 so I think it is that.
> Can you try again ?
Still fails.
Note to self: a simple test is "curl http://www.memotoo.com/syncmldev"
Can you try again ?
With IPv4, I get a redirect. With IPv6, it just hangs.
1:18 a.m.
On Fr, 2013-11-22 at 09:35 +0100, Thomas Pequet wrote:
Le 21/11/2013 20:39, Patrick Ohly a écrit :
> On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
>> Le 21/11/2013 17:32, Patrick Ohly a écrit :
>>
>>> On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
>>>> Do you use IPv6 ?
>>> Could be. At least the machine and the transparent proxy it sits behind
>>> are IPv6 enabled. I'm less sure whether libsoup uses IPv6.
>> Ok I havejust change the IPv6 so I think it is that.
>> Can you try again ?
> Still fails.
>
> Note to self: a simple test is "curl http://www.memotoo.com/syncmldev"
Can you try again ?
Now it seems to work, at least with curl. Still need to run SyncML
tests.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
6:12 a.m.
On Fri, 2013-11-22 at 15:18 +0100, Patrick Ohly wrote:
On Fr, 2013-11-22 at 09:35 +0100, Thomas Pequet wrote:
> Le 21/11/2013 20:39, Patrick Ohly a écrit :
> > On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
> >> Le 21/11/2013 17:32, Patrick Ohly a écrit :
> >>
> >>> On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
> >>>> Do you use IPv6 ?
> >>> Could be. At least the machine and the transparent proxy it sits
behind
> >>> are IPv6 enabled. I'm less sure whether libsoup uses IPv6.
> >> Ok I havejust change the IPv6 so I think it is that.
> >> Can you try again ?
> > Still fails.
> >
> > Note to self: a simple test is "curl
http://www.memotoo.com/syncmldev"
> Can you try again ?
Now it seems to work, at least with curl. Still need to run SyncML
tests.
Now it seems to be broken again.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
10:06 p.m.
Le 23/11/2013 20:12, Patrick Ohly a écrit :
On Fri, 2013-11-22 at 15:18 +0100, Patrick Ohly wrote:
> On Fr, 2013-11-22 at 09:35 +0100, Thomas Pequet wrote:
>> Le 21/11/2013 20:39, Patrick Ohly a écrit :
>>> On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
>>>> Le 21/11/2013 17:32, Patrick Ohly a écrit :
>>>>
>>>>> On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
>>>>>> Do you use IPv6 ?
>>>>> Could be. At least the machine and the transparent proxy it sits
behind
>>>>> are IPv6 enabled. I'm less sure whether libsoup uses IPv6.
>>>> Ok I havejust change the IPv6 so I think it is that.
>>>> Can you try again ?
>>> Still fails.
>>>
>>> Note to self: a simple test is "curl
http://www.memotoo.com/syncmldev"
>> Can you try again ?
> Now it seems to work, at least with curl. Still need to run SyncML
> tests.
Now it seems to be broken again.
Very strange ...
Can you send me the log or tcpdump you have ?
8:55 p.m.
On Sun, 2013-11-24 at 12:06 +0100, Thomas Pequet wrote:
Le 23/11/2013 20:12, Patrick Ohly a écrit :
> On Fri, 2013-11-22 at 15:18 +0100, Patrick Ohly wrote:
>> On Fr, 2013-11-22 at 09:35 +0100, Thomas Pequet wrote:
>>> Le 21/11/2013 20:39, Patrick Ohly a écrit :
>>>> On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
>>>>> Le 21/11/2013 17:32, Patrick Ohly a écrit :
>>>>>
>>>>>> On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
>>>>>>> Do you use IPv6 ?
>>>>>> Could be. At least the machine and the transparent proxy it sits
behind
>>>>>> are IPv6 enabled. I'm less sure whether libsoup uses IPv6.
>>>>> Ok I havejust change the IPv6 so I think it is that.
>>>>> Can you try again ?
>>>> Still fails.
>>>>
>>>> Note to self: a simple test is "curl
http://www.memotoo.com/syncmldev"
>>> Can you try again ?
>> Now it seems to work, at least with curl. Still need to run SyncML
>> tests.
> Now it seems to be broken again.
>
Very strange ...
It seems the failure is not perfectly deterministic. It now failed once,
worked the next second, then failed again.
Can you send me the log or tcpdump you have ?
I can't take a tcpdump. The actual machine doing the hostname lookup and
IPv6 lookup is an HTTP proxy that I don't have access to.
If you ping me on IRC as #pohly on freenode.org or irc.gnome.org, I can
run the curl test command while you run a tcpdump on your end, perhaps
using a test port or something else that doesn't interfere with
production.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
9:06 p.m.
On Mon, 2013-11-25 at 10:55 +0100, Patrick Ohly wrote:
On Sun, 2013-11-24 at 12:06 +0100, Thomas Pequet wrote:
> Le 23/11/2013 20:12, Patrick Ohly a écrit :
> > On Fri, 2013-11-22 at 15:18 +0100, Patrick Ohly wrote:
> >> On Fr, 2013-11-22 at 09:35 +0100, Thomas Pequet wrote:
> >>> Le 21/11/2013 20:39, Patrick Ohly a écrit :
> >>>> On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
> >>>>> Le 21/11/2013 17:32, Patrick Ohly a écrit :
> >>>>>
> >>>>>> On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
> >>>>>>> Do you use IPv6 ?
> >>>>>> Could be. At least the machine and the transparent proxy it
sits behind
> >>>>>> are IPv6 enabled. I'm less sure whether libsoup uses
IPv6.
> >>>>> Ok I havejust change the IPv6 so I think it is that.
> >>>>> Can you try again ?
> >>>> Still fails.
> >>>>
> >>>> Note to self: a simple test is "curl
http://www.memotoo.com/syncmldev"
> >>> Can you try again ?
> >> Now it seems to work, at least with curl. Still need to run SyncML
> >> tests.
> > Now it seems to be broken again.
> >
> Very strange ...
It seems the failure is not perfectly deterministic. It now failed once,
worked the next second, then failed again.
Perhaps it's also a load issue.
$ time curl -v http://www.memotoo.com/syncmldev
[eventually the redirect is returned]
real 2m1.055s
user 0m0.008s
sys 0m0.000s
2 minutes for a single request? It's not surprising that there are
timeouts, most likely in the HTTP proxy.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
9:07 p.m.
On Mon, 2013-11-25 at 11:06 +0100, Patrick Ohly wrote:
On Mon, 2013-11-25 at 10:55 +0100, Patrick Ohly wrote:
> On Sun, 2013-11-24 at 12:06 +0100, Thomas Pequet wrote:
> > Le 23/11/2013 20:12, Patrick Ohly a écrit :
> > > On Fri, 2013-11-22 at 15:18 +0100, Patrick Ohly wrote:
> > >> On Fr, 2013-11-22 at 09:35 +0100, Thomas Pequet wrote:
> > >>> Le 21/11/2013 20:39, Patrick Ohly a écrit :
> > >>>> On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
> > >>>>> Le 21/11/2013 17:32, Patrick Ohly a écrit :
> > >>>>>
> > >>>>>> On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet
wrote:
> > >>>>>>> Do you use IPv6 ?
> > >>>>>> Could be. At least the machine and the transparent
proxy it sits behind
> > >>>>>> are IPv6 enabled. I'm less sure whether libsoup
uses IPv6.
> > >>>>> Ok I havejust change the IPv6 so I think it is that.
> > >>>>> Can you try again ?
> > >>>> Still fails.
> > >>>>
> > >>>> Note to self: a simple test is "curl
http://www.memotoo.com/syncmldev"
> > >>> Can you try again ?
> > >> Now it seems to work, at least with curl. Still need to run SyncML
> > >> tests.
> > > Now it seems to be broken again.
> > >
> > Very strange ...
>
> It seems the failure is not perfectly deterministic. It now failed once,
> worked the next second, then failed again.
Perhaps it's also a load issue.
$ time curl -v http://www.memotoo.com/syncmldev
[eventually the redirect is returned]
real 2m1.055s
user 0m0.008s
sys 0m0.000s
2 minutes for a single request? It's not surprising that there are
timeouts, most likely in the HTTP proxy.
... on the other hand, it is always fast on my desktop machine. <shrug>
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
10:26 p.m.
Le 25/11/2013 11:07, Patrick Ohly a écrit :
On Mon, 2013-11-25 at 11:06 +0100, Patrick Ohly wrote:
> On Mon, 2013-11-25 at 10:55 +0100, Patrick Ohly wrote:
>> On Sun, 2013-11-24 at 12:06 +0100, Thomas Pequet wrote:
>>> Le 23/11/2013 20:12, Patrick Ohly a écrit :
>>>> On Fri, 2013-11-22 at 15:18 +0100, Patrick Ohly wrote:
>>>>> On Fr, 2013-11-22 at 09:35 +0100, Thomas Pequet wrote:
>>>>>> Le 21/11/2013 20:39, Patrick Ohly a écrit :
>>>>>>> On Do, 2013-11-21 at 18:14 +0100, Thomas Pequet wrote:
>>>>>>>> Le 21/11/2013 17:32, Patrick Ohly a écrit :
>>>>>>> >>>>>>>>>
On Do, 2013-11-21 at 17:10 +0100, Thomas Pequet wrote:
>>>>>>>>>> Do you use IPv6 ?
>>>>>>>>> Could be. At least the machine and the transparent
proxy it sits behind
>>>>>>>>> are IPv6 enabled. I'm less sure whether libsoup
uses IPv6.
>>>>>>>> Ok I havejust change the IPv6 so I think it is that.
>>>>>>>> Can you try again ?
>>>>>>> Still fails.
>>>>>> >>>>>>> Note to
self: a simple test is "curl http://www.memotoo.com/syncmldev"
>>>>>> Can you try again ?
>>>>> Now it seems to work, at least with curl. Still need to run SyncML
>>>>> tests.
>>>> Now it seems to be broken again.
>>> >>> Very strange ...
>> It seems the failure is not perfectly deterministic. It now failed once,
>> worked the next second, then failed again.
> Perhaps it's also a load issue.
> $ time curl -v http://www.memotoo.com/syncmldev
> [eventually the redirect is returned]
> real 2m1.055s
> user 0m0.008s
> sys 0m0.000s
> 2 minutes for a single request? It's not surprising
that there are
> timeouts, most likely in the HTTP proxy.
... on the other hand, it is always fast on my desktop machine. <shrug>
OK
the problem is the IPv6 because she was assigned with a DHCPv6.
I think now it is OK.
10:46 p.m.
On Mon, 2013-11-25 at 12:26 +0100, Thomas Pequet wrote:
Le 25/11/2013 11:07, Patrick Ohly a écrit :
> On Mon, 2013-11-25 at 11:06 +0100, Patrick Ohly wrote:
>> $ time curl -v http://www.memotoo.com/syncmldev
>> [eventually the redirect is returned]
>> real 2m1.055s
>> user 0m0.008s
>> sys 0m0.000s
>>
>> 2 minutes for a single request? It's not surprising that there are
>> timeouts, most likely in the HTTP proxy.
> ... on the other hand, it is always fast on my desktop machine. <shrug>
OK the problem is the IPv6 because she was assigned with a DHCPv6.
I think now it is OK.
Yes, now curl worked quickly five times in a row.
--
Best Regards, Patrick Ohly
The content of this message is my personal opinion only and although
I am an employee of Intel, the statements I make here in no way
represent Intel's position on the issue, nor am I authorized to speak
on behalf of Intel on this matter.
3219
days inactive
3223
days old
syncevolution@synevolution.org
13 comments
2 participants
participants (2)
-
Patrick Ohly -
Thomas Pequet