Hello,
I am working on a communication project where we use ice4j, specifically ICE
and PseudoTCP. We have detected and fixed a few bugs which arise in our
usage scenario. How should I proceed to give the fixes back to the
community?
Thx Jakub
Hey Jakub,
Good to hear from you. You can just send the patches here, to this mailing list.
You would also need to sign the BJ contributor agreement as explained
on jitsi.org:
https://jitsi.org/commitaccess and
https://jitsi.org/faq/patch
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
···
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hello,
I am working on a communication project where we use ice4j, specifically ICE
and PseudoTCP. We have detected and fixed a few bugs which arise in our
usage scenario. How should I proceed to give the fixes back to the
community?Thx Jakub
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server
. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method
. ice4j-keep-alive-15-seconds - changes the keep-alive interval from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property
. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nomination
Please tell me about the next steps.
Jakub
ice4j-fix-TURN.patch (1.56 KB)
ice4j-keep-alive-15-seconds.patch (427 Bytes)
ice4j-logging-improvements.patch (8.26 KB)
ice4j-no-keep-alives.patch (960 Bytes)
ice4j-pseudotcp.patch (7.2 KB)
ice4j-unnecessary-nomination.patch (535 Bytes)
···
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hey Jakub,
Good to hear from you. You can just send the patches here, to this mailing
list.
You would also need to sign the BJ contributor agreement as explained on
jitsi.org:
<https://jitsi.org/commitaccess> https://jitsi.org/commitaccess and
<https://jitsi.org/faq/patch> https://jitsi.org/faq/patch
The contributor agreement itself is available here:
<http://bluejimp.com/bca.pdf> http://bluejimp.com/bca.pdf
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek < <mailto:jakub.adamek@ishisystems.com> jakub.adamek@ishisystems.com> wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
<mailto:emcho@jitsi.org> emcho@jitsi.org PHONE:
+33.1.77.62.43.30
<https://jitsi.org> https://jitsi.org FAX:
+33.1.77.62.47.31
Hi Jakub,
Thank you for you patches !
I've started reviewing them and have 2 qestions regards the pseudo TCP one:
1. In PseudoTcpSocketImpl you've added a call setOption(SO_TIMEOUT, 0);
public PseudoTcpSocketImpl(long conv_id, DatagramSocket sock)
{
pseudoTcp = new PseudoTCPBase(this, conv_id);
//Default MTU
setMTU(1450);
this.socket = sock;
try {
setOption(SO_TIMEOUT, 0);
} catch (SocketException e) {
throw new RuntimeException(e);
}
}
Any comments that could describe why this call is required ? Is it
used by the system or is causing a timeout exception if not called ?
2. In PseudoTcpSocketImpl:
private void checkOpenAndCreate(boolean create)
throws SocketException
{
if (isClosed())
{
throw new SocketException("Socket is closed");
}
if (!create)
{
if (!isConnected())
{
throw new SocketException("Socket is not connected");
}
}
}
This method is called only once with create=true from
connect(SocketAddress,int).
I moved only the isClosed() check there and removed the method. Do you
agree with that ?
Other changes seem to be fine, although there were some missing java
docs and few formatting fixes required.
Regards,
Pawel
···
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
· ice4j-fix-TURN – important fix when receiving data from a TURN
server· ice4j-pseudotcp – fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method· ice4j-keep-alive-15-seconds – changes the keep-alive interval from
3s to 15s to better align with ICE RFC· ice4j-no-keep-alives – allows to switch off keep-alives using
system property· ice4j-logging-improvements – several minor logging improvements
· ice4j-unnecessary-nomination – minor fix, skips unnecessary
nominationPlease tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixesHey Jakub,
Good to hear from you. You can just send the patches here, to this mailing
list.You would also need to sign the BJ contributor agreement as explained on
jitsi.org:https://jitsi.org/commitaccess and
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek <jakub.adamek@ishisystems.com> > wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi Jakub,
On more question this time about system property that disables the keep alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't
it contain our package name in front like "org.ice4j.NoKeepAlives" to
avoid conflicts ? Do you agree ?
Regards,
Pawel
···
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
· ice4j-fix-TURN – important fix when receiving data from a TURN
server· ice4j-pseudotcp – fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method· ice4j-keep-alive-15-seconds – changes the keep-alive interval from
3s to 15s to better align with ICE RFC· ice4j-no-keep-alives – allows to switch off keep-alives using
system property· ice4j-logging-improvements – several minor logging improvements
· ice4j-unnecessary-nomination – minor fix, skips unnecessary
nominationPlease tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixesHey Jakub,
Good to hear from you. You can just send the patches here, to this mailing
list.You would also need to sign the BJ contributor agreement as explained on
jitsi.org:https://jitsi.org/commitaccess and
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek <jakub.adamek@ishisystems.com> > wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi Pawel,
Yes, I agree, and the name of the constant should not contain the PROPERTY_
prefix to be more similar to e.g. StackProperties.MAX_CTRAN_RETRANSMISSIONS.
We could perhaps even place it into StackProperties.
Jakub
···
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of
Pawel Domas
Sent: Tuesday, September 17, 2013 7:52 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On more question this time about system property that disables the keep
alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't it
contain our package name in front like "org.ice4j.NoKeepAlives" to avoid
conflicts ? Do you agree ?
Regards,
Pawel
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nominationPlease tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixesHey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.You would also need to sign the BJ contributor agreement as explained
on
jitsi.org:https://jitsi.org/commitaccess and
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek > <jakub.adamek@ishisystems.com> > wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi Pawel,
I am not sure any more, why this was necessary. I have done several
different experiments with PseudoTcp. I propose to skip this change for now
and I will come again if I find it is necessary and why.
Jakub
···
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of
Pawel Domas
Sent: Tuesday, September 17, 2013 6:48 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
Thank you for you patches !
I've started reviewing them and have 2 qestions regards the pseudo TCP one:
1. In PseudoTcpSocketImpl you've added a call setOption(SO_TIMEOUT, 0);
public PseudoTcpSocketImpl(long conv_id, DatagramSocket sock)
{
pseudoTcp = new PseudoTCPBase(this, conv_id);
//Default MTU
setMTU(1450);
this.socket = sock;
try {
setOption(SO_TIMEOUT, 0);
} catch (SocketException e) {
throw new RuntimeException(e);
}
}
Any comments that could describe why this call is required ? Is it used by
the system or is causing a timeout exception if not called ?
2. In PseudoTcpSocketImpl:
private void checkOpenAndCreate(boolean create)
throws SocketException
{
if (isClosed())
{
throw new SocketException("Socket is closed");
}
if (!create)
{
if (!isConnected())
{
throw new SocketException("Socket is not connected");
}
}
}
This method is called only once with create=true from
connect(SocketAddress,int).
I moved only the isClosed() check there and removed the method. Do you agree
with that ?
Other changes seem to be fine, although there were some missing java docs
and few formatting fixes required.
Regards,
Pawel
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nominationPlease tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixesHey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.You would also need to sign the BJ contributor agreement as explained
on
jitsi.org:https://jitsi.org/commitaccess and
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek > <jakub.adamek@ishisystems.com> > wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi Jakub,
Hi Pawel,
Yes, I agree, and the name of the constant should not contain the PROPERTY_
prefix to be more similar to e.g. StackProperties.MAX_CTRAN_RETRANSMISSIONS.
We could perhaps even place it into StackProperties.Jakub
I moved it to StackProperties. All patches are now applied and acked
on our contributors page:
https://jitsi.org/Development/TeamAndContributors
Thanks for patches once gain !
Regards,
Pawel
···
On Tue, Sep 17, 2013 at 9:22 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of
Pawel Domas
Sent: Tuesday, September 17, 2013 7:52 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixesHi Jakub,
On more question this time about system property that disables the keep
alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't it
contain our package name in front like "org.ice4j.NoKeepAlives" to avoid
conflicts ? Do you agree ?Regards,
PawelOn Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek <jakub.adamek@ishisystems.com> > wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nominationPlease tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixesHey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.You would also need to sign the BJ contributor agreement as explained
on
jitsi.org:https://jitsi.org/commitaccess and
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek >> <jakub.adamek@ishisystems.com> >> wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi Pawel,
Thank you, I have updated my project to use the committed version.
I like the way you cleaned up the patches.
Jakub
···
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of
Pawel Domas
Sent: Wednesday, September 18, 2013 5:51 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On Tue, Sep 17, 2013 at 9:22 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hi Pawel,
Yes, I agree, and the name of the constant should not contain the
PROPERTY_ prefix to be more similar to e.g.
StackProperties.MAX_CTRAN_RETRANSMISSIONS.
We could perhaps even place it into StackProperties.
Jakub
I moved it to StackProperties. All patches are now applied and acked on our
contributors page:
https://jitsi.org/Development/TeamAndContributors
Thanks for patches once gain !
Regards,
Pawel
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf
Of Pawel Domas
Sent: Tuesday, September 17, 2013 7:52 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixesHi Jakub,
On more question this time about system property that disables the
keep alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't
it contain our package name in front like "org.ice4j.NoKeepAlives" to
avoid conflicts ? Do you agree ?Regards,
PawelOn Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek > <jakub.adamek@ishisystems.com> > wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nominationPlease tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixesHey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.You would also need to sign the BJ contributor agreement as explained
on
jitsi.org:https://jitsi.org/commitaccess and
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek >> <jakub.adamek@ishisystems.com> >> wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few
bugswhich arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi all, hi Pawel,
I am here with a few new patches:
. ice4j-fix-deadlock fixes an important deadlock in PseudoTCP where
the read_notify lock might conflict with the "synchronized" on
PseudoTCPBase.recv
. ice4j-logging-improvement improves logging and fixes reaction on
InterruptedException
. ice4j-skip-unnecessary-bind speeds up the ICE process in case
where you have a local network interface unable to reach Internet at all (in
our case, it was a VirtualBox network)
o this last one was made by Viral Patel, whom I include as CC
Have a nice day,
Jakub
ice4j-fix-deadlock.patch (3.41 KB)
ice4j-logging-improvements.patch (2.07 KB)
ice4j-skip-unnecessary-bind.patch (817 Bytes)
···
-----Original Message-----
From: Jakub Adamek
Sent: Wednesday, September 18, 2013 11:51 AM
To: 'dev@jitsi.org'
Subject: RE: [jitsi-dev] FW: ice4j fixes
Hi Pawel,
Thank you, I have updated my project to use the committed version.
I like the way you cleaned up the patches.
Jakub
-----Original Message-----
From: <mailto:dev-bounces@jitsi.org> dev-bounces@jitsi.org [
<mailto:dev-bounces@jitsi.org> mailto:dev-bounces@jitsi.org] On Behalf Of
Pawel Domas
Sent: Wednesday, September 18, 2013 5:51 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On Tue, Sep 17, 2013 at 9:22 PM, Jakub Adamek < <mailto:jakub.adamek@ishisystems.com> jakub.adamek@ishisystems.com> wrote:
Hi Pawel,
Yes, I agree, and the name of the constant should not contain the
PROPERTY_ prefix to be more similar to e.g.
StackProperties.MAX_CTRAN_RETRANSMISSIONS.
We could perhaps even place it into StackProperties.
Jakub
I moved it to StackProperties. All patches are now applied and acked on our
contributors page:
<https://jitsi.org/Development/TeamAndContributors>
https://jitsi.org/Development/TeamAndContributors
Thanks for patches once gain !
Regards,
Pawel
-----Original Message-----
From: <mailto:dev-bounces@jitsi.org> dev-bounces@jitsi.org [
<mailto:dev-bounces@jitsi.org> mailto:dev-bounces@jitsi.org] On Behalf
Of Pawel Domas
Sent: Tuesday, September 17, 2013 7:52 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On more question this time about system property that disables the
keep alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't
it contain our package name in front like "org.ice4j.NoKeepAlives" to
avoid conflicts ? Do you agree ?
Regards,
Pawel
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek > < <mailto:jakub.adamek@ishisystems.com> jakub.adamek@ishisystems.com> > wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server
. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method
. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property
. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nomination
Please tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [ <mailto:emcho@jitsi.org> mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.
You would also need to sign the BJ contributor agreement as explained
on
<https://jitsi.org/commitaccess> https://jitsi.org/commitaccess and
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek >> < <mailto:jakub.adamek@ishisystems.com> jakub.adamek@ishisystems.com> >> wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few
bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
<mailto:emcho@jitsi.org> emcho@jitsi.org PHONE:
+33.1.77.62.43.30
+33.1.77.62.47.31
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
<mailto:dev@jitsi.org> dev@jitsi.org
Unsubscribe instructions and other list options:
<http://lists.jitsi.org/mailman/listinfo/dev>
http://lists.jitsi.org/mailman/listinfo/dev
Hi Jakub,
Hi all, hi Pawel,
I am here with a few new patches:
· ice4j-fix-deadlock fixes an important deadlock in PseudoTCP where
the read_notify lock might conflict with the “synchronized” on
PseudoTCPBase.recv· ice4j-logging-improvement improves logging and fixes reaction on
InterruptedException· ice4j-skip-unnecessary-bind speeds up the ICE process in case
where you have a local network interface unable to reach Internet at all (in
our case, it was a VirtualBox network)o this last one was made by Viral Patel, whom I include as CC
Have a nice day,
Jakub
Thanks for the patches !
I tried to apply deadlock fix, but now one test is failing. I don't
know yet why, but I'm looking into it.
Regards,
Pawel
···
On Fri, Nov 8, 2013 at 5:17 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
-----Original Message-----
From: Jakub Adamek
Sent: Wednesday, September 18, 2013 11:51 AM
To: 'dev@jitsi.org'
Subject: RE: [jitsi-dev] FW: ice4j fixesHi Pawel,
Thank you, I have updated my project to use the committed version.
I like the way you cleaned up the patches.
Jakub
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of
Pawel Domas
Sent: Wednesday, September 18, 2013 5:51 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On Tue, Sep 17, 2013 at 9:22 PM, Jakub Adamek <jakub.adamek@ishisystems.com> > > wrote:
Hi Pawel,
Yes, I agree, and the name of the constant should not contain the
PROPERTY_ prefix to be more similar to e.g.
StackProperties.MAX_CTRAN_RETRANSMISSIONS.
We could perhaps even place it into StackProperties.
Jakub
I moved it to StackProperties. All patches are now applied and acked on our
contributors page:
https://jitsi.org/Development/TeamAndContributors
Thanks for patches once gain !
Regards,
Pawel
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf
Of Pawel Domas
Sent: Tuesday, September 17, 2013 7:52 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On more question this time about system property that disables the
keep alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't
it contain our package name in front like "org.ice4j.NoKeepAlives" to
avoid conflicts ? Do you agree ?
Regards,
Pawel
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek > >> <jakub.adamek@ishisystems.com> > >> wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server
. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method
. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property
. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nomination
Please tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.
You would also need to sign the BJ contributor agreement as explained
on
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek > >>> <jakub.adamek@ishisystems.com> > >>> wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few
bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi Jakub,
I've commited your patches, but unfortunately I do not feel confident
enough in ice area to commit Viral's patch. Someone who is more
experienced in it should do this.
Regards,
Pawel
···
On Fri, Nov 8, 2013 at 5:17 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hi all, hi Pawel,
I am here with a few new patches:
· ice4j-fix-deadlock fixes an important deadlock in PseudoTCP where
the read_notify lock might conflict with the “synchronized” on
PseudoTCPBase.recv· ice4j-logging-improvement improves logging and fixes reaction on
InterruptedException· ice4j-skip-unnecessary-bind speeds up the ICE process in case
where you have a local network interface unable to reach Internet at all (in
our case, it was a VirtualBox network)o this last one was made by Viral Patel, whom I include as CC
Have a nice day,
Jakub
-----Original Message-----
From: Jakub Adamek
Sent: Wednesday, September 18, 2013 11:51 AM
To: 'dev@jitsi.org'
Subject: RE: [jitsi-dev] FW: ice4j fixesHi Pawel,
Thank you, I have updated my project to use the committed version.
I like the way you cleaned up the patches.
Jakub
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of
Pawel Domas
Sent: Wednesday, September 18, 2013 5:51 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On Tue, Sep 17, 2013 at 9:22 PM, Jakub Adamek <jakub.adamek@ishisystems.com> > > wrote:
Hi Pawel,
Yes, I agree, and the name of the constant should not contain the
PROPERTY_ prefix to be more similar to e.g.
StackProperties.MAX_CTRAN_RETRANSMISSIONS.
We could perhaps even place it into StackProperties.
Jakub
I moved it to StackProperties. All patches are now applied and acked on our
contributors page:
https://jitsi.org/Development/TeamAndContributors
Thanks for patches once gain !
Regards,
Pawel
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf
Of Pawel Domas
Sent: Tuesday, September 17, 2013 7:52 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On more question this time about system property that disables the
keep alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't
it contain our package name in front like "org.ice4j.NoKeepAlives" to
avoid conflicts ? Do you agree ?
Regards,
Pawel
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek > >> <jakub.adamek@ishisystems.com> > >> wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server
. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method
. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property
. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nomination
Please tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.
You would also need to sign the BJ contributor agreement as explained
on
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek > >>> <jakub.adamek@ishisystems.com> > >>> wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few
bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi,
I think I've found the problem with the latest patch. Here's what I think happens:
1. [Thread 1] The line read = pseudoTcp.recv(buffer, offset, length) in PseudoTcpSocketImpl returns 0.
2. [Thread 2] Some new data comes in and triggers read_notify.notifyAll().
3. [Thread 1] This thread is still unaware of the new data and decides to call read_notify.wait(left) and gets stuck there since there's no data coming in that can trigger read_notify.notifyAll().
I have attached a patch that should address this issue.
Best Regards
Carl Hasselskog
PseudoTCP-deadlock.patch (1.5 KB)
···
-----Original Message-----
From: Paweł Domas [mailto:pawel.domas@jitsi.org]
Sent: den 19 november 2013 15:13
To: Carl Hasselskog
Subject: Fwd: [jitsi-dev] FW: ice4j fixes
---------- Forwarded message ----------
From: Paweł Domas <pawel.domas@jitsi.org>
Date: Mon, Nov 18, 2013 at 6:56 PM
Subject: Re: [jitsi-dev] FW: ice4j fixes
To: Jitsi Developers <dev@jitsi.org>
Hi Jakub,
On Fri, Nov 8, 2013 at 5:17 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hi all, hi Pawel,
I am here with a few new patches:
· ice4j-fix-deadlock fixes an important deadlock in PseudoTCP where
the read_notify lock might conflict with the “synchronized” on
PseudoTCPBase.recv· ice4j-logging-improvement improves logging and fixes reaction on
InterruptedException· ice4j-skip-unnecessary-bind speeds up the ICE process in case
where you have a local network interface unable to reach Internet at
all (in our case, it was a VirtualBox network)o this last one was made by Viral Patel, whom I include as CC
Have a nice day,
Jakub
Thanks for the patches !
I tried to apply deadlock fix, but now one test is failing. I don't know yet why, but I'm looking into it.
Regards,
Pawel
-----Original Message-----
From: Jakub Adamek
Sent: Wednesday, September 18, 2013 11:51 AM
To: 'dev@jitsi.org'
Subject: RE: [jitsi-dev] FW: ice4j fixesHi Pawel,
Thank you, I have updated my project to use the committed version.
I like the way you cleaned up the patches.
Jakub
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf
OfPawel Domas
Sent: Wednesday, September 18, 2013 5:51 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On Tue, Sep 17, 2013 at 9:22 PM, Jakub Adamek > <jakub.adamek@ishisystems.com> > > wrote:
Hi Pawel,
Yes, I agree, and the name of the constant should not contain the
PROPERTY_ prefix to be more similar to e.g.
StackProperties.MAX_CTRAN_RETRANSMISSIONS.
We could perhaps even place it into StackProperties.
Jakub
I moved it to StackProperties. All patches are now applied and acked
on ourcontributors page:
https://jitsi.org/Development/TeamAndContributors
Thanks for patches once gain !
Regards,
Pawel
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf
Of Pawel Domas
Sent: Tuesday, September 17, 2013 7:52 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On more question this time about system property that disables the
keep alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't
it contain our package name in front like "org.ice4j.NoKeepAlives" to
avoid conflicts ? Do you agree ?
Regards,
Pawel
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek > >> <jakub.adamek@ishisystems.com> > >> wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server
. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method
. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property
. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nomination
Please tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.
You would also need to sign the BJ contributor agreement as
explainedon
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek > >>> <jakub.adamek@ishisystems.com> > >>> wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few
bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
Hi Carl,
Hi,
I think I've found the problem with the latest patch. Here's what I think happens:1. [Thread 1] The line read = pseudoTcp.recv(buffer, offset, length) in PseudoTcpSocketImpl returns 0.
2. [Thread 2] Some new data comes in and triggers read_notify.notifyAll().
3. [Thread 1] This thread is still unaware of the new data and decides to call read_notify.wait(left) and gets stuck there since there's no data coming in that can trigger read_notify.notifyAll().I have attached a patch that should address this issue.
Best Regards
Carl Hasselskog
Yes, indeed this must be the case. Now the tests run fine. Thanks for
the patch I'll commit it soon 
Pawel
···
On Wed, Nov 20, 2013 at 10:43 AM, Carl Hasselskog <carl@degoo.com> wrote:
-----Original Message-----
From: Paweł Domas [mailto:pawel.domas@jitsi.org]
Sent: den 19 november 2013 15:13
To: Carl Hasselskog
Subject: Fwd: [jitsi-dev] FW: ice4j fixes---------- Forwarded message ----------
From: Paweł Domas <pawel.domas@jitsi.org>
Date: Mon, Nov 18, 2013 at 6:56 PM
Subject: Re: [jitsi-dev] FW: ice4j fixes
To: Jitsi Developers <dev@jitsi.org>Hi Jakub,
On Fri, Nov 8, 2013 at 5:17 PM, Jakub Adamek <jakub.adamek@ishisystems.com> wrote:
Hi all, hi Pawel,
I am here with a few new patches:
· ice4j-fix-deadlock fixes an important deadlock in PseudoTCP where
the read_notify lock might conflict with the “synchronized” on
PseudoTCPBase.recv· ice4j-logging-improvement improves logging and fixes reaction on
InterruptedException· ice4j-skip-unnecessary-bind speeds up the ICE process in case
where you have a local network interface unable to reach Internet at
all (in our case, it was a VirtualBox network)o this last one was made by Viral Patel, whom I include as CC
Have a nice day,
Jakub
Thanks for the patches !
I tried to apply deadlock fix, but now one test is failing. I don't know yet why, but I'm looking into it.
Regards,
Pawel-----Original Message-----
From: Jakub Adamek
Sent: Wednesday, September 18, 2013 11:51 AM
To: 'dev@jitsi.org'
Subject: RE: [jitsi-dev] FW: ice4j fixesHi Pawel,
Thank you, I have updated my project to use the committed version.
I like the way you cleaned up the patches.
Jakub
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf
OfPawel Domas
Sent: Wednesday, September 18, 2013 5:51 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On Tue, Sep 17, 2013 at 9:22 PM, Jakub Adamek >> <jakub.adamek@ishisystems.com> >> >> wrote:
Hi Pawel,
Yes, I agree, and the name of the constant should not contain the
PROPERTY_ prefix to be more similar to e.g.
StackProperties.MAX_CTRAN_RETRANSMISSIONS.
We could perhaps even place it into StackProperties.
Jakub
I moved it to StackProperties. All patches are now applied and acked
on ourcontributors page:
https://jitsi.org/Development/TeamAndContributors
Thanks for patches once gain !
Regards,
Pawel
-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf
Of Pawel Domas
Sent: Tuesday, September 17, 2013 7:52 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hi Jakub,
On more question this time about system property that disables the
keep alives.
I'm not sure if using so general name "NoKeepAlives" is ok. Shouldn't
it contain our package name in front like "org.ice4j.NoKeepAlives" to
avoid conflicts ? Do you agree ?
Regards,
Pawel
On Mon, Sep 9, 2013 at 7:05 PM, Jakub Adamek >> >>> <jakub.adamek@ishisystems.com> >> >>> wrote:
Hey,
I have cleaned up and prepared the patches. Kindly find them attached:
. ice4j-fix-TURN - important fix when receiving data from a TURN
server
. ice4j-pseudotcp - fixes PseudoTCP on Android, improves logging,
and fixes a serious bug in the int read() method
. ice4j-keep-alive-15-seconds - changes the keep-alive interval
from
3s to 15s to better align with ICE RFC
. ice4j-no-keep-alives - allows to switch off keep-alives using
system property
. ice4j-logging-improvements - several minor logging improvements
. ice4j-unnecessary-nomination - minor fix, skips unnecessary
nomination
Please tell me about the next steps.
Jakub
-----Original Message-----
From: Emil Ivov [mailto:emcho@jitsi.org]
Sent: Thursday, August 15, 2013 1:24 PM
To: Jitsi Developers
Cc: Jakub Adamek
Subject: Re: [jitsi-dev] FW: ice4j fixes
Hey Jakub,
Good to hear from you. You can just send the patches here, to this
mailing list.
You would also need to sign the BJ contributor agreement as
explainedon
The contributor agreement itself is available here:
You can sign, scan and send this one directly to me.
Please ping me if you have any questions.
Cheers,
Emil
On Wed, Aug 14, 2013 at 8:39 PM, Jakub Adamek >> >>>> <jakub.adamek@ishisystems.com> >> >>>> wrote:
Hello,
I am working on a communication project where we use ice4j,
specifically ICE and PseudoTCP. We have detected and fixed a few
bugs
which arise in our usage scenario. How should I proceed to give the
fixes back to the community?
Thx Jakub
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
--
Emil Ivov, Ph.D. 67000 Strasbourg,
Project Lead France
Jitsi
emcho@jitsi.org PHONE: +33.1.77.62.43.30
https://jitsi.org FAX: +33.1.77.62.47.31
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
_______________________________________________
dev mailing list
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev
_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev