[jitsi-dev] Jitsi's memory usage


#1

Hi,

My Jitsi is currenty idling Jitsi with about 20 contacts in roster (with photos). My Jitsi currently occupies 380 MB of my RAM ... right now more than any other process on my Win7.

Users often ask me about this. I know Jitsi is Java based but even then 380 MB seems really high. Is there any reason for this ... or could this be optimized somehow?

Cheers,
Richard

···

________________________________

Prohlášení: k tomuto e-mailu je připojeno prohlášení, jehož
plné znění naleznete na stránce http://www.ipex.cz/prohlaseni


#2

Hey Richard,

We haven't investigated so not sure what the reasons are.

Regular use should be below or around 250 so this does seem a bit high.

Emil

···

On 23.05.13, 12:36, Richard Karmazín wrote:

Hi,

My Jitsi is currenty idling Jitsi with about 20 contacts in roster (with
photos). My Jitsi currently occupies 380 MB of my RAM … right now more
than any other process on my Win7.

Users often ask me about this. I know Jitsi is Java based but even then
380 MB seems really high. Is there any reason for this … or could this
be optimized somehow?

Cheers,

Richard

------------------------------------------------------------------------

Prohlášení: k tomuto e-mailu je připojeno prohlášení, jehož
plné znění naleznete na stránce http://www.ipex.cz/prohlaseni

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev

--
https://jitsi.org


#3

is this related

maybe related: this user reported that jitsi is very memory-hungry (~700
mb) on his machine. on my machine jitsi currently uses ~174 mb - less
than firefox but more than thunderbird.

Yes, this is caused by the bug in the JRE native code (it allocates memory
for each NIC with an IPV6 index of 0). As Jitsi queries the up-status of a
NIC quite often, I'm not surprised about this memory usage.

Ingo

is net.dll responsible for this? is it possible to replace net.dll? this
could possibly solve the 'multiple NIC' issue, which is currently a
knock out and 'our people' are continuing using skype simply because it
works.

···

Am 14.05.2013 11:42, schrieb Ingo Bauersachs:

Am 23.05.2013 11:39, schrieb Emil Ivov:

Hey Richard,

We haven't investigated so not sure what the reasons are.

Regular use should be below or around 250 so this does seem a bit high.

Emil

On 23.05.13, 12:36, Richard Karmazín wrote:

Hi,

My Jitsi is currenty idling Jitsi with about 20 contacts in roster (with
photos). My Jitsi currently occupies 380 MB of my RAM … right now more
than any other process on my Win7.

Users often ask me about this. I know Jitsi is Java based but even then
380 MB seems really high. Is there any reason for this … or could this
be optimized somehow?

Cheers,

Richard

------------------------------------------------------------------------

Prohlášení: k tomuto e-mailu je připojeno prohlášení, jehož
plné znění naleznete na stránce http://www.ipex.cz/prohlaseni

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev


#4

The leak deascribed above is in net.dll, yes. Net.dll is from the JRE and hence cannot easily be replaced. One would have to get the source from the OpenJDK, patch the IPv6 bug and then compile it - and keep it up to date with every new JRE release.

Maybe someone could construct a DoS with a Java-Applet that constantly queries the network status? This would possibly give Oracle an incentive to get active...

Not that the memory usage described by Richard is not necessarily related.

Ingo

···

Am 14.05.2013 11:42, schrieb Ingo Bauersachs:

maybe related: this user reported that jitsi is very memory-hungry (~700
mb) on his machine. on my machine jitsi currently uses ~174 mb - less
than firefox but more than thunderbird.

Yes, this is caused by the bug in the JRE native code (it allocates memory
for each NIC with an IPV6 index of 0). As Jitsi queries the up-status of a
NIC quite often, I'm not surprised about this memory usage.

Ingo

is net.dll responsible for this? is it possible to replace net.dll? this
could possibly solve the 'multiple NIC' issue, which is currently a
knock out and 'our people' are continuing using skype simply because it
works.


#5

well ,mayby it's not necessary to replace net.dll but create a
'jitsi-net.dll' providing the necessary funtionality - just my thoughts

···

Am 24.05.2013 08:52, schrieb Ingo Bauersachs:

Am 14.05.2013 11:42, schrieb Ingo Bauersachs:

maybe related: this user reported that jitsi is very memory-hungry (~700
mb) on his machine. on my machine jitsi currently uses ~174 mb - less
than firefox but more than thunderbird.

Yes, this is caused by the bug in the JRE native code (it allocates memory
for each NIC with an IPV6 index of 0). As Jitsi queries the up-status of a
NIC quite often, I'm not surprised about this memory usage.

Ingo

is net.dll responsible for this? is it possible to replace net.dll? this
could possibly solve the 'multiple NIC' issue, which is currently a
knock out and 'our people' are continuing using skype simply because it
works.

The leak deascribed above is in net.dll, yes. Net.dll is from the JRE and hence cannot easily be replaced. One would have to get the source from the OpenJDK, patch the IPv6 bug and then compile it - and keep it up to date with every new JRE release.

Maybe someone could construct a DoS with a Java-Applet that constantly queries the network status? This would possibly give Oracle an incentive to get active...

Not that the memory usage described by Richard is not necessarily related.

Ingo

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev


#6

Btw: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6458027

From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of
g.spellauge
Sent: Freitag, 24. Mai 2013 08:59
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

well ,mayby it's not necessary to replace net.dll but create a
'jitsi-net.dll' providing the necessary funtionality - just my thoughts

>>>> maybe related: this user reported that jitsi is very memory-hungry

(~700

>>>> mb) on his machine. on my machine jitsi currently uses ~174 mb - less
>>>> than firefox but more than thunderbird.
>>> Yes, this is caused by the bug in the JRE native code (it allocates
memory
>>> for each NIC with an IPV6 index of 0). As Jitsi queries the up-status

of

a
>>> NIC quite often, I'm not surprised about this memory usage.
>>>
>>> Ingo
>>>
>> is net.dll responsible for this? is it possible to replace net.dll?

this

>> could possibly solve the 'multiple NIC' issue, which is currently a
>> knock out and 'our people' are continuing using skype simply because it
>> works.
> The leak deascribed above is in net.dll, yes. Net.dll is from the JRE

and

hence cannot easily be replaced. One would have to get the source from the
OpenJDK, patch the IPv6 bug and then compile it - and keep it up to date

with

every new JRE release.
>
> Maybe someone could construct a DoS with a Java-Applet that constantly
queries the network status? This would possibly give Oracle an incentive

to

get active...
>
> Not that the memory usage described by Richard is not necessarily

related.

···

-----Original Message-----
Am 24.05.2013 08:52, schrieb Ingo Bauersachs:
>> Am 14.05.2013 11:42, schrieb Ingo Bauersachs:
>
> Ingo
>
>
> _______________________________________________
> 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


#7

thanks for this link.

we would like to try to set -Djava.net.preferIPv4Stack=true property but
are unsure how and where to do this because in the windows environment
jitsi is startet by jitsi.exe - could we just append
-Djava.net.preferIPv4Stack=true property to the command?

···

Am 24.05.2013 09:43, schrieb Ingo Bauersachs:

Btw: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6458027

-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of
g.spellauge
Sent: Freitag, 24. Mai 2013 08:59
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

well ,mayby it's not necessary to replace net.dll but create a
'jitsi-net.dll' providing the necessary funtionality - just my thoughts

Am 24.05.2013 08:52, schrieb Ingo Bauersachs:

Am 14.05.2013 11:42, schrieb Ingo Bauersachs:

maybe related: this user reported that jitsi is very memory-hungry

(~700

mb) on his machine. on my machine jitsi currently uses ~174 mb - less
than firefox but more than thunderbird.

Yes, this is caused by the bug in the JRE native code (it allocates

memory

for each NIC with an IPV6 index of 0). As Jitsi queries the up-status

of

a

NIC quite often, I'm not surprised about this memory usage.

Ingo

is net.dll responsible for this? is it possible to replace net.dll?

this

could possibly solve the 'multiple NIC' issue, which is currently a
knock out and 'our people' are continuing using skype simply because it
works.

The leak deascribed above is in net.dll, yes. Net.dll is from the JRE

and

hence cannot easily be replaced. One would have to get the source from the
OpenJDK, patch the IPv6 bug and then compile it - and keep it up to date

with

every new JRE release.

Maybe someone could construct a DoS with a Java-Applet that constantly

queries the network status? This would possibly give Oracle an incentive

to

get active...

Not that the memory usage described by Richard is not necessarily

related.

Ingo

_______________________________________________
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


#8

I am happy to see discussion flowing :slight_smile:

Would love to try -Djava.net.preferIPv4Stack=true so +1 to g.spellauge. Is there a way in jitsi.exe to pass arguments to Java VM? I think it would be usefull for debugging/testing.

Richard

···

-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of g.spellauge
Sent: Friday, May 24, 2013 10:19 AM
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

thanks for this link.

we would like to try to set -Djava.net.preferIPv4Stack=true property but are unsure how and where to do this because in the windows environment jitsi is startet by jitsi.exe - could we just append -Djava.net.preferIPv4Stack=true property to the command?

Am 24.05.2013 09:43, schrieb Ingo Bauersachs:

Btw: http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6458027

-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf
Of g.spellauge
Sent: Freitag, 24. Mai 2013 08:59
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

well ,mayby it's not necessary to replace net.dll but create a
'jitsi-net.dll' providing the necessary funtionality - just my
thoughts

Am 24.05.2013 08:52, schrieb Ingo Bauersachs:

Am 14.05.2013 11:42, schrieb Ingo Bauersachs:

maybe related: this user reported that jitsi is very
memory-hungry

(~700

mb) on his machine. on my machine jitsi currently uses ~174 mb -
less than firefox but more than thunderbird.

Yes, this is caused by the bug in the JRE native code (it
allocates

memory

for each NIC with an IPV6 index of 0). As Jitsi queries the
up-status

of

a

NIC quite often, I'm not surprised about this memory usage.

Ingo

is net.dll responsible for this? is it possible to replace net.dll?

this

could possibly solve the 'multiple NIC' issue, which is currently a
knock out and 'our people' are continuing using skype simply
because it works.

The leak deascribed above is in net.dll, yes. Net.dll is from the
JRE

and

hence cannot easily be replaced. One would have to get the source
from the OpenJDK, patch the IPv6 bug and then compile it - and keep
it up to date

with

every new JRE release.

Maybe someone could construct a DoS with a Java-Applet that
constantly

queries the network status? This would possibly give Oracle an
incentive

to

get active...

Not that the memory usage described by Richard is not necessarily

related.

Ingo

_______________________________________________
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

________________________________

Prohlášení: k tomuto e-mailu je připojeno prohlášení, jehož
plné znění naleznete na stránce http://www.ipex.cz/prohlaseni


#9

Would love to try -Djava.net.preferIPv4Stack=true so +1 to g.spellauge. Is
there a way in jitsi.exe to pass arguments to Java VM?

This possibility somehow got lost when we migrated to the embedded loading
of the JVM. run.c [1] would have to be patched. Any volunteers? :slight_smile:

I think it would be usefull for debugging/testing.

Richard

Ingo

[1] https://github.com/jitsi/jitsi/blob/master/src/native/windows/run/run.c


#10

executing Jitsi.exe -h from the comandline displays

Jitsi 2.2.4603.9615
age: jitsi [OPTIONS] [uri-to-call]

-c, --config=DIR use DIR for config files
-d, --debug print debugging messages to stdout
-h, --help display this help message and exit
-m, --multiple do not ensure single instance
-6, --ipv6 prefer IPv6 addresses where possible only
-4, --ipv4 forces use of IPv4 only
-v, --version display the current version and exit

is -4 the option we are looking for i.e. is it equivalent to the jre
option -Djava.net.preferIPv4Stack=true ?

···

Am 24.05.2013 11:59, schrieb Ingo Bauersachs:

Would love to try -Djava.net.preferIPv4Stack=true so +1 to g.spellauge. Is
there a way in jitsi.exe to pass arguments to Java VM?

This possibility somehow got lost when we migrated to the embedded loading
of the JVM. run.c [1] would have to be patched. Any volunteers? :slight_smile:

I think it would be usefull for debugging/testing.

Richard

Ingo

[1] https://github.com/jitsi/jitsi/blob/master/src/native/windows/run/run.c

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev


#11

executing Jitsi.exe -h from the comandline displays

Jitsi 2.2.4603.9615
age: jitsi [OPTIONS] [uri-to-call]

-c, --config=DIR use DIR for config files
-d, --debug print debugging messages to stdout
-h, --help display this help message and exit
-m, --multiple do not ensure single instance
-6, --ipv6 prefer IPv6 addresses where possible only
-4, --ipv4 forces use of IPv4 only
-v, --version display the current version and exit

is -4 the option we are looking for i.e. is it equivalent to the jre
option -Djava.net.preferIPv4Stack=true ?

Yes. It internally sets the preferIPv4Stack property to true. I forgot about
those options...

Ingo


#12

hello,

we just checked with one of our 'problem-computers': switched order of
interfaces until jitsi was unable to initiate a voice chat, added -4 to
the comandline, started jitsi and had no problem establishing a voice
connection.

richard, could you please verify that this will solve your issue as well.

···

Am 24.05.2013 13:51, schrieb Ingo Bauersachs:

executing Jitsi.exe -h from the comandline displays

Jitsi 2.2.4603.9615
age: jitsi [OPTIONS] [uri-to-call]

-c, --config=DIR use DIR for config files
-d, --debug print debugging messages to stdout
-h, --help display this help message and exit
-m, --multiple do not ensure single instance
-6, --ipv6 prefer IPv6 addresses where possible only
-4, --ipv4 forces use of IPv4 only
-v, --version display the current version and exit

is -4 the option we are looking for i.e. is it equivalent to the jre
option -Djava.net.preferIPv4Stack=true ?

Yes. It internally sets the preferIPv4Stack property to true. I forgot about
those options...

Ingo

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev


#13

After trying -4 parameter for few days i am still not sure it helped. Jitsi's current memory usage is 317MB which is slightly better (from 370MB), but still not ideal.

Richard

···

-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of g.spellauge
Sent: Friday, May 24, 2013 2:00 PM
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

hello,

we just checked with one of our 'problem-computers': switched order of interfaces until jitsi was unable to initiate a voice chat, added -4 to the comandline, started jitsi and had no problem establishing a voice connection.

richard, could you please verify that this will solve your issue as well.

Am 24.05.2013 13:51, schrieb Ingo Bauersachs:

executing Jitsi.exe -h from the comandline displays

Jitsi 2.2.4603.9615
age: jitsi [OPTIONS] [uri-to-call]

-c, --config=DIR use DIR for config files
-d, --debug print debugging messages to stdout
-h, --help display this help message and exit
-m, --multiple do not ensure single instance
-6, --ipv6 prefer IPv6 addresses where possible only
-4, --ipv4 forces use of IPv4 only
-v, --version display the current version and exit

is -4 the option we are looking for i.e. is it equivalent to the jre
option -Djava.net.preferIPv4Stack=true ?

Yes. It internally sets the preferIPv4Stack property to true. I forgot
about those options...

Ingo

_______________________________________________
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

________________________________

Prohlášení: k tomuto e-mailu je připojeno prohlášení, jehož
plné znění naleznete na stránce http://www.ipex.cz/prohlaseni


#14

After trying -4 parameter for few days i am still not sure it helped. Jitsi's current memory usage is 317MB which is slightly better (from 370MB), but still not ideal.

What do you consider to be the RAM capacity of your average user?

While 317MB may be more than other applications it probably wouldn't
lead to any noticeable problems on most modern boxes.

Emil

···

On 28.05.13, 12:18, Richard Karmazín wrote:

Richard

-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of g.spellauge
Sent: Friday, May 24, 2013 2:00 PM
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

hello,

we just checked with one of our 'problem-computers': switched order of interfaces until jitsi was unable to initiate a voice chat, added -4 to the comandline, started jitsi and had no problem establishing a voice connection.

richard, could you please verify that this will solve your issue as well.

Am 24.05.2013 13:51, schrieb Ingo Bauersachs:

executing Jitsi.exe -h from the comandline displays

Jitsi 2.2.4603.9615
age: jitsi [OPTIONS] [uri-to-call]

-c, --config=DIR use DIR for config files
-d, --debug print debugging messages to stdout
-h, --help display this help message and exit
-m, --multiple do not ensure single instance
-6, --ipv6 prefer IPv6 addresses where possible only
-4, --ipv4 forces use of IPv4 only
-v, --version display the current version and exit

is -4 the option we are looking for i.e. is it equivalent to the jre
option -Djava.net.preferIPv4Stack=true ?

Yes. It internally sets the preferIPv4Stack property to true. I forgot
about those options...

Ingo

_______________________________________________
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

________________________________

Prohlášení: k tomuto e-mailu je připojeno prohlášení, jehož
plné znění naleznete na stránce http://www.ipex.cz/prohlaseni

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev

--
https://jitsi.org


#15

Considering my Jitsi instance is currently sitting at over 1.5Gb memory usage I'm very interested in this discussion

I'll try the -4 option

···

On 28/05/2013 7:31 PM, Emil Ivov wrote:

On 28.05.13, 12:18, Richard Karmaz�n wrote:

After trying -4 parameter for few days i am still not sure it helped. Jitsi's current memory usage is 317MB which is slightly better (from 370MB), but still not ideal.

What do you consider to be the RAM capacity of your average user?

While 317MB may be more than other applications it probably wouldn't
lead to any noticeable problems on most modern boxes.

Emil

Richard

-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of g.spellauge
Sent: Friday, May 24, 2013 2:00 PM
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

hello,

we just checked with one of our 'problem-computers': switched order of interfaces until jitsi was unable to initiate a voice chat, added -4 to the comandline, started jitsi and had no problem establishing a voice connection.

richard, could you please verify that this will solve your issue as well.

Am 24.05.2013 13:51, schrieb Ingo Bauersachs:

executing Jitsi.exe -h from the comandline displays

Jitsi 2.2.4603.9615
age: jitsi [OPTIONS] [uri-to-call]

-c, --config=DIR use DIR for config files
-d, --debug print debugging messages to stdout
-h, --help display this help message and exit
-m, --multiple do not ensure single instance
-6, --ipv6 prefer IPv6 addresses where possible only
-4, --ipv4 forces use of IPv4 only
-v, --version display the current version and exit

is -4 the option we are looking for i.e. is it equivalent to the jre
option -Djava.net.preferIPv4Stack=true ?

Yes. It internally sets the preferIPv4Stack property to true. I forgot
about those options...

Ingo

_______________________________________________
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

________________________________

Prohl�en�: k tomuto e-mailu je p�ipojeno prohl�en�, jeho�
pln� zn�n� naleznete na str�nce http://www.ipex.cz/prohlaseni

_______________________________________________
dev mailing list
dev@jitsi.org
Unsubscribe instructions and other list options:
http://lists.jitsi.org/mailman/listinfo/dev


#16

You are certainly not the only one.
I was looking for a softphone for a thin-client.
Already found that jitsi cannot work any more on systems (XP, Ubuntu) with 1GB.
(If the memory consumption continues, I would suggest to rename the project: obesitel. ;-))

···

-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of Craig Chandler
Sent: Thursday, May 30, 2013 11:57 PM
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

Considering my Jitsi instance is currently sitting at over 1.5Gb memory
usage I'm very interested in this discussion

I'll try the -4 option

On 28/05/2013 7:31 PM, Emil Ivov wrote:

On 28.05.13, 12:18, Richard Karmazín wrote:

After trying -4 parameter for few days i am still not sure it helped. Jitsi's current memory usage is 317MB which is slightly better (from 370MB), but still not ideal.

What do you consider to be the RAM capacity of your average user?

While 317MB may be more than other applications it probably wouldn't
lead to any noticeable problems on most modern boxes.

Emil

Richard

-----Original Message-----
From: dev-bounces@jitsi.org [mailto:dev-bounces@jitsi.org] On Behalf Of g.spellauge
Sent: Friday, May 24, 2013 2:00 PM
To: Jitsi Developers
Subject: Re: [jitsi-dev] Jitsi's memory usage

hello,

we just checked with one of our 'problem-computers': switched order of interfaces until jitsi was unable to initiate a voice chat, added -4 to the comandline, started jitsi and had no problem establishing a voice connection.

richard, could you please verify that this will solve your issue as well.

Am 24.05.2013 13:51, schrieb Ingo Bauersachs:

executing Jitsi.exe -h from the comandline displays

Jitsi 2.2.4603.9615
age: jitsi [OPTIONS] [uri-to-call]

-c, --config=DIR use DIR for config files
-d, --debug print debugging messages to stdout
-h, --help display this help message and exit
-m, --multiple do not ensure single instance
-6, --ipv6 prefer IPv6 addresses where possible only
-4, --ipv4 forces use of IPv4 only
-v, --version display the current version and exit

is -4 the option we are looking for i.e. is it equivalent to the jre
option -Djava.net.preferIPv4Stack=true ?

Yes. It internally sets the preferIPv4Stack property to true. I forgot
about those options...

Ingo

_______________________________________________
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

________________________________

Prohlášení: k tomuto e-mailu je připojeno prohlášení, jehož
plné znění naleznete na stránce http://www.ipex.cz/prohlaseni

_______________________________________________
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

______________________________________________________________________
Dit bericht kan informatie bevatten die niet voor u is bestemd. Indien u niet de geadresseerde bent of dit bericht abusievelijk aan u is toegezonden, wordt u verzocht dat aan de afzender te melden en het bericht te verwijderen. De Staat aanvaardt geen aansprakelijkheid voor schade, van welke aard ook, die verband houdt met risico's verbonden aan het electronisch verzenden van berichten.

This message may contain information that is not intended for you. If you are not the addressee or if this message was sent to you by mistake, you are requested to inform the sender and delete the message. The State accepts no liability for damage of any kind resulting from the risks inherent in the electronic transmission of messages.


#17

I don't find obesity to be a particularly funny subject nor do I find this mail particularly constructive.

I already mentioned that:

We (the developer) are aware of the memory consumption that Jitsi has. We (BlueJimp) are currently working on issues that we find more prioritary (related to general stability, Jitsi Videobridge, and Jitsi for Android).

If anyone else would like to take a stab on optimizing memory consumption: by all means.

In the mean time, the kind of bashing that we are seeing here is really not helping anyone in any way.

Emil

···

On 31.05.13, 10:50, J.Witvliet@mindef.nl wrote:

You are certainly not the only one.
I was looking for a softphone for a thin-client.
Already found that jitsi cannot work any more on systems (XP, Ubuntu) with 1GB.
(If the memory consumption continues, I would suggest to rename the project: obesitel. ;-))

--
https://jitsi.org