VoIP Mailing List Archives

[krice at suspicious.org]

Upgrade to trunk... Many many issues have been resolved since 1.0.1 was the
current release

[anthony.minessale at g...]

Also remember,
Actually completely uninstall and erase /usr/local/freeswitch and the 1.0.1 source tree and freshly install the new one.
If you try to upgrade on top of a release with trunk it will cause more problems for you.


On Wed, Jan 28, 2009 at 3:11 AM, Ken Rice <krice@suspicious.org (krice@suspicious.org)> wrote:

[msc at freeswitch.org]

FYI,

I added a note on the reporting bugs page that specifically talks
about this scenario:
http://wiki.freeswitch.org/wiki/Reporting_Bugs#Upgrading_From_Older_Revisions

-MC

On Wed, Jan 28, 2009 at 6:01 AM, Anthony Minessale
<anthony.minessale@gmail.com> wrote:

[pmhshz at gmail.com]

Hi Anthony,

I found interesting result while testing Freeswitch, and it might be cause
of freezing out of freeswitch,

I updated my system (as you told) with latest stable version Freeswitch
1.0.2
First of all I set sps to 100,
Then I sends call approximately 100 per seconds, Freeswitch works fine and
handles all the calls very well.

After that I send 130 calls per seconds, and magic happen now, Freeswitch
handles first 100 calls only.
all the preceding calls were failed (even not appeared in freeswitch console
why?)

When I put ngrep trace, System responds with 503 Maximum Calls In Progress.
(as below)
###########################################################
#
U FSFSFSFSFS -> GWGWGWGWGW
SIP/2.0 503 Maximum Calls In Progress.
Via: SIP/2.0/UDP GWGWGWGWGW;branch=z9hG4bK53eafabb;rport=5060.
From: "99999" <sip:99999@GWGWGWGWGW>;tag=as2e10c170.
To: <sip:1234567890@FSFSFSFSFS >;tag=K3jSUFrDHpmmB.
Call-ID: 0feb229e58afe7be17110deb361dc234@GWGWGWGWGW.
CSeq: 102 INVITE.
Retry-After: 300.
User-Agent: FreeSWITCH-mod_sofia/1.0.2-exported.
Accept: application/sdp.
Allow: INVITE, ACK, BYE, CANCEL, OPTIONS, PRACK, MESSAGE, SUBSCRIBE, NOTIFY,
REFER, UPDATE, REGISTER, INFO, PUBLISH.
Supported: timer, precondition, path, replaces.
Allow-Events: talk, presence, dialog, call-info, sla,
include-session-description, presence.winfo, message-summary, refer.
Content-Length: 0.
.
#####################################################################


Now another issue to note down is that,
After all above happened and active calls comes to zero,
I just make a single call which also fails with response 503 - Maximum Calls
In Progress.


Is this intended behaviour, should I increase SPS to overcome this. or
something like bug.

Please let me know what should be the resolution for this.

Thanks,
msp



Anthony Minessale-2 wrote:

[brian at freeswitch.org]

You might also want to set max-proceeding on the sofia profile.

/b

On Jan 29, 2009, at 10:06 AM, shehzad p wrote:

[anthony.minessale at g...]

When you get it in that state what do you see when you execute

fsctl sps

is the sps a very low number?

Did the sps drop by itself from the value you originally set it to?

Are you using 32 bit?

if so try all of these commands in your shell before starting FS

[pmhshz at gmail.com]

Thanks, Anthony

In my previous test sps did not changed,
but in recent test sps was dropped to 0 itself (as below).
===============================================================
UP 0 years, 0 days, 5 hours, 1 minute, 53 seconds, 878 milliseconds, 190
microseconds
5474 session(s) since startup
75 session(s) 0/0
=============================================================

My system is 32 bit.
CPU is Intel(R) Xeon(R) CPU X3220 @ 2.40GHz
And RAM is 4GB

Output of ulimit -a is:
ulimit -a: (set after first test)
core file size (blocks, -c) unlimited
data seg size (kbytes, -d) unlimited
max nice (-e) 20
file size (blocks, -f) unlimited
pending signals (-i) unlimited
max locked memory (kbytes, -l) unlimited
max memory size (kbytes, -m) unlimited
open files (-n) 999999
pipe size (512 bytes, -p) 8
POSIX message queues (bytes, -q) unlimited
max rt priority (-r) unlimited
stack size (kbytes, -s) 244
cpu time (seconds, -t) unlimited
max user processes (-u) unlimited
virtual memory (kbytes, -v) unlimited
file locks (-x) unlimited
===================================================


BTW using trunk on production system is safe?

Warm thanks for kind responses...



Anthony Minessale-2 wrote:

[pmhshz at gmail.com]

When freeswitch freezes, we can't connect to it to check sps status,
but once we were able to connect and at that time it was showing 0/0 sps.

thanks...

shehzad p wrote:

[anthony.minessale at g...]

if you are using unix you can use the supplied script

scripts/freeswitch-gcore

to capture a copy of the resident memory and I can have a look perhaps.

Trunk is safe for production as we are in beta stage for a release of 1.0.3 at this time.



On Fri, Jan 30, 2009 at 9:29 AM, shehzad p <pmhshz@gmail.com (pmhshz@gmail.com)> wrote:

[pmhshz at gmail.com]

Hi Anthony,

Freeswitch 1.0.2 was crashed on last test again...
BT is on http://pastebin.freeswitch.org/6979.

I tried to use scripts/freeswitch-gcore, to capture resident memory, but
before the command complete the process system hanged up so only half output
was captured.< http://www.nabble.com/file/p21761523/gcore-fs.txt
gcore-fs.txt >


Now I have checkout from trunk and will post back if any new thing found...

Thanks,
msp

Anthony Minessale-2 wrote:

[pmhshz at gmail.com]

Hi Anthoy,

I just tested the trunk-11575,

In one test FS served for 1 Hour and 10 minutes and processed about 28600
sesisons.
When it get crashed there was 244 active sessions and cps was 5/200

Here is full BT attached http://www.nabble.com/file/p21765660/fs_bt_full.txt
fs_bt_full.txt

Thanks,
msp

Anthony Minessale-2 wrote:

[anthony.minessale at g...]

Clearly you have an issue with your javascript code.

You have the Garbage collector blocking in every thread.

Are you doing any endless loops in your code where you do not check session.ready() as a condition for
continuing the script?

any time session.ready() fails you must immediately exit.

Are you using session.execute to execute long blocking operations like bridging many calls or entering a conference?
You should avoid doing this as all the collective scripts on the system share a common Garbage Collector provided by the
JS engine and it can lead to the exact issues you describe if the code is not properly designed.

What else does you script do that are things provided by FS such as playing files and executing applications.



On Sat, Jan 31, 2009 at 3:44 AM, shehzad p <pmhshz@gmail.com (pmhshz@gmail.com)> wrote:

[pmhshz at gmail.com]

Hi Anthony,

How can I handle Garbage collector related things. Is there any way to look
out where js variables are causing some kind of fault related to Garbage
collector?

There are some loops but they finish in hardly two or three iterations.

I didn't check session.ready() before/after fetching values from database
(odbc), which might block until response from server comes back. might this
cause fault as you said?

Thanks for help...




Anthony Minessale-2 wrote:

[pmhshz at gmail.com]

Hi anthony,

There are so many crash occured, on FS 1.0.1 Server, all of them were same
(means BT was same)
but this last one was looking totally different so I feel that it should be
posted...
http://www.nabble.com/file/p21775256/bt_full_new.txt bt_full_new.txt


shehzad p wrote:

[brian at freeswitch.org]

I highly recommend you upgrade to 1.03RC1 or SVN Trunk. Chances are we have already fixed these issues.

/b

On Feb 1, 2009, at 7:35 AM, shehzad p wrote: