Hello All,
Trying to see if someone can point me in the right direction with an issue I'm having. We are currently experiencing issues with call quality and latency when we have 15 concurrent calls while trying to dial other numbers for waiting agents. If we keep it under 10 we see no issues at all. Our current environment is the following:
Server 2012 R2 x64 6GB Single Processor
Dialogic HMP 3.0 Update 349
SonicWALL NSA 2600
Any assistance is greatly appreciated.
Hello Everyone! Hope this the correct forum to post an issue we have encounter in our environment. The following is what we are running:
1. Server 2012 R2 (VM on ESXi-6.5.0) - HMP 3.0 Service Update 349 -
Issue is we are noticing call quality and latency when we have 15 concurrent connected calls while trying to dial other numbers for waiting agents. If we keep it below 10, then we have no issues. If any of you have encounter an issue like this could you point me in the right direction on how I can remediate this. Thanks.
Hi Greg,
Can you elaborate in more detail with respect to quality and latency in this case? Meaning is this only specific to audio heard on the call? Or are there other metrics, like how long it takes to process events at the application level.
Just curious?
The first thing I would check is to make sure the system HW support HPET and that is enabled in the VM as well:
www.dialogic.com/.../pm_hmp_win
Jeff
Hi Jeff,
Thank you so much for replying. What we have noticed is that the IP based dialers should be able to handle 100-120 people each but if we have more than 25 combined we get call quality and network connectivity issues that affect even remote sites. If the volume of calls increase the internal web servers slow down so much it make agents wait between screens. If we exceed 30-35 users the mitel pbx which is hosted outside of the firewall but on the same network will even start to drop calls and decrease in call quality.
Hope this helps on understanding the issue we are having. Again thank you for your reply. Much Appreciated!
Jeff,
Just an update. Followed the article you sent and the results is the following:
"HPET is Hardware & BIOS supported and is enabled on this system."
If only a single processor, do you really mean single core assigned? That may be that system is underpowered in this case. What is the speed.
Generally speaking I think you may need at least 4 cores in this case assigned in that case as a BKM. There is some info on sizing in the release guide, but I don't know the last time that was changed (table 1):
www.dialogic.com/.../release_guide.pdf
Jeff
Hi Jeff,
Thanks for the reply. Here is a screenshot of the resources the VM has:
This is the Host Resources:
We are attempting to use the latest HMP SU 382 but cannot get past gc_Start with any of the multiple targets. We have a good valid license and have tried reapplying the license/and restoring defaults to no avail. Reverting to any previous SU “fixes” the issue:
04/09/2018 14:35:15.393 6852 6932 sm_main.cpp:340 ! 0 ! gc_h3r:>> h3r_Start : gc_start_structp->version=513, media_operational_mode=EmbeddedMedia
04/09/2018 14:35:15.393 6852 6932 sm.cpp:3778 ! 0 ! gc_h3r:>> Sending cmd [4]=>0XD1A10518 count [0]=>0XD1A1051C
04/09/2018 14:35:15.393 6852 6932 sm.cpp:3801 ! 0 ! gc_h3r:<< Received response [0]=>0XD1A1051C count [24]=>0XD1A10504
04/09/2018 14:35:15.393 6852 6932 sm.cpp:3778 ! 0 ! gc_h3r:>> Sending cmd [1]=>0XD1A1051D count [-1]=>0X2E5EFAE3
04/09/2018 14:35:15.393 6852 6932 sm.cpp:3801 ! 0 ! gc_h3r:<< Received response [-4]=>0X2E5EFAE0 count [-1]=>0X2E5EFAE3
04/09/2018 14:35:15.393 6852 6932 gc_h3r ERR1 sm.cpp:3805 ! 0 ! Error -4 communicating with License Server
04/09/2018 14:35:15.393 6852 6932 gc_h3r ERR1 sm.cpp:2906 ! 0 ! << SharonMgr::performLicenceAndParameterValidation: Could not perform checkout Feature not supported or FlexLM license file is not active
04/09/2018 14:35:15.393 6852 6932 gc_h3r ERR1 sm.cpp:2351 ! 0 ! IPCCLIB Licence Validation Failed 
.
04/09/2018 14:35:15.901 6852 6932 gc_h3r ERR1 sm_main.cpp:356 ! 0 ! >> h3r_Start : caught exception 6 while creating Sharon Manager
04/09/2018 14:35:15.901 6852 6932 gc ERR1 gcprod ----- _gp_StartAllCCLibs() - Library libgch3r.dll start procedure returned error: 29
04/09/2018 14:35:15.901 6852 6932 gc ERR1 gclib ::::> gc_Start() - _gp_StartAllCCLibs() failed:-1
Actually, the issue is pretty clear in this case with the DATE header which is the 12th line:
Date: Wed, 28 Mar 2018 14:20:41 EDT
Review of RFC3261 specification states that this behavior is not allowed as per section listed below:
20.17 Date
The Date header field contains the date and time. Unlike HTTP/1.1,
SIP only supports the most recent RFC 1123 [20] format for dates. As
in [H3.3], SIP restricts the time zone in SIP-date to "GMT", while
RFC 1123 allows any time zone. An RFC 1123 date is case-sensitive.
The Date header field reflects the time when the request or response
is first sent.
Thus the sender end-point or gateway is not following the RFC
Jeff
We have a cisco based environment where a “Call-Info:” call header is being included and causing HMP to generate a SIP Parse error instead of handling the call. In the same environment a secondary set of SIP (checked side by side)without the Call-Info: header works fine but the addition is required by other items/applications present in this alternative environment. What could be the cause as the header format looks ok to me?
400 SIP Parser Error : Syntax Error, line 12, column 27 of the header value
Via: SIP/2.0/TCP 10.6.72.108:5060;branch=z9hG4bKa2ZIV4f5ru2JKxPM97Vxxg~~42985
To: <sip:4265153@10.6.70.22;transport=tcp>;tag=5e721c0-1646060a-13c4-65014-8143-1ca75f85-8143
From: 4143343965 <sip:4143343965@10.6.72.108:5060>;tag=ds9be0e78b
CSeq: 1 INVITE
Content-Length: 0
Call-ID: A7FE5F3031D311E8895ECA7F2BB929E9-15222511024454011@10.6.72.108
03/30/2018 09:40:00.892 5632 2712 sip_stack Error 00000A98 ERROR - PARSER - SIP Parser Error : Syntax Error, line 12, column 27 of the header value
03/30/2018 09:40:00.892 5632 2712 sip_stack Error 00000A98 ERROR - PARSER - SipParserStartParsing - Syntax Error - The parser failed to parse line 12
INVITE sip:4265152@10.6.70.22;transport=tcp SIP/2.0
Via: SIP/2.0/TCP 10.6.72.108:5060;branch=z9hG4bKa2ZIV4f5ru2JKxPM97Vxxg~~43042
Max-Forwards: 68
To: <sip:4265152@10.6.70.22;transport=tcp>
From: 4143343965 <sip:4143343965@10.6.72.108:5060>;tag=dse2df5bac
Call-ID: 8ECB826C31EB11E88BB2CA7F2BB929E9-15222613556334024@10.6.72.108
CSeq: 1 INVITE
Content-Length: 288
Contact: <sip:4143343965@10.6.72.108:5060;transport=tcp>
Expires: 60
User-Agent: CVP 11.5 (1) ES-1 Build-349
Date: Wed, 28 Mar 2018 14:20:41 EDT
Min-SE: 3600
Cisco-Guid: 2395701868-0837489128-2343750271-0733555177
Allow: INVITE, OPTIONS, BYE, CANCEL, ACK, PRACK, UPDATE, REFER, SUBSCRIBE, NOTIFY, INFO, REGISTER
Allow-Events: telephone-event
P-Asserted-Identity: <sip:4143343965@10.6.72.10>
Session-Expires: 3600
Content-Disposition: session;handling=required
Call-Info: <sip:10.6.72.10>;purpose=x-cisco-origIP
Cisco-Gucid: 8ECB826C31EB11E88BB2CA7F2BB929E9
Supported: rel100
Supported: timer
Supported: resource-priority
Supported: replaces
Supported: sdp-anat
Content-Type: application/sdp
User-To-User: 8ECB826C31EB11E88BB2CA7F2BB929E9;210024053609;4143343965;8002721325;IVR
App-Info: <10.6.72.108:7000:7443>
v=0
o=Cisco-CVP-B2BUA 289844526 289844527 IN IP4 10.6.72.108
s=SIP Call
c=IN IP4 10.6.72.10
t=0 0
m=audio 22722 RTP/AVP 0 100 101
c=IN IP4 10.6.72.10
a=rtpmap:0 PCMU/8000
a=rtpmap:100 X-NSE/8000
a=fmtp:100 192-194
a=rtpmap:101 telephone-event/8000
a=fmtp:101 0-15
a=ptime:20
Hello,
I have a 1PCC application using HMP Windows. We are interfacing to a PBX which requires that all SIP messages generated by HMP include a User-Agent header.
I set a User-Agent header using gc_SetUserInfo() as described in section 4.9 of the GlobalCall IP for HMP Technology guide:
char * ua = "User-Agent: Support";
gc_util_insert_parm_ref(&gcParmBlk,IPSET_SIP_MSGINFO,IPPARM_SIP_HDR,(unsigned char)strlen(ua)+1,ua);
But not all messages generated from HMP show the User-Agent header.
On incoming calls (Invite-OK-ACK), the OK response from HMP includes the User agent. This works fine.
On outgoing calls (Invite-OK-ACK), the Invite from HMP includes the user agent, but the ACK does not. This causes the PBX to ignore the Ack and call setup fails.
Is there a way to add the User-Agent header to the ACK?
Thanks,
Tony
Thanks Jeff,
We were using gc_setUserInfo on the line device with the GC_ALLCALLS paramter, but that did not apply the User-Agent header on the ack. We also tried doing gc_setUserInfo just prior to gc_makecall with the GC_SINGLE_SIP_SESSION paramter, but that didn't work either.
Do you have any other suggestions? We are interfacing to a PBX that requires that user-agents be whitelisted, and they are ignoring our ACK's because of the lack of User Agent header.
Tony
Ok, if you have 6 cores assigned that should be ok in this case. However, I noticed that you are on very old version:
Dialogic HMP 3.0 Update 349 (released July 2013 )
We didn't add support for VM 6.x till SU367 as per release update notes:
www.dialogic.com/.../release_update.pdf
It might be worth upgrading in this case to be on a version of which we testing with VMware of that revision. Current revision if SU382 release earlier this year. Also for the fact newer SU contains many improvements at host library and firmware levels which you don't have that could aid in this case (ie updates in code branches, etc).
Regards,
Jeff
This could be do to some changes that were made in parameter checking when the switch was made to VS 2015 base. Please check at IP_VIRTBOARD parameters (like total calls, max calls, etc), contain the correct counts to match that of the license in place and not use the default settings.
Jeff
Hi:
I have a problem with the resources of NIC, my virtual machine detects that the resources are not fully in the Manager devices, by the way, in the DCM (Configuration Manager) I can't see the virtual board DM3 HMP. the technology of virtualization is AWS (Amazon Web Services). My question is:
Are there any problem of compatibility between Dialogic HMP 3.0 and technology VMWare Cloud AWS (Amazon Web Services)?
Unfortunately in this case AWS has not been qualified for Windows HMP 3.0 baseline as of yet. Its only been used for different PowerMedia XMS product which is Linux based.
However, since we do support VMware in general as a Virtual Machine software, I suspect it could be something with the actual VM platform configuration provided by AWS in this case where HPET needs to be enabled at physical server level and passed thru the to the VM configuration instance. I guess you would have to contact AWS on the likelihood of getting that done if possible.
Regards,
Jeff
Linux hmp - I'd like to create a program to listen to a channel that has an active call going and record the call. I want the call to proceed normally with its prompts, dtmf, Asr, etc., without interrupting it. So it seems like the logical choice would be to fire up a background process to listen in and record everything. Is this possible?
Record a 1 legged call. I want to record a normal call, which would only have 1 TS, however functions like mreciottdata require two timeslots since they are designed to record a two party call. Can someone steer me to a solution to record just a normal call? I appears that I can't open the channel from another process, but I'm unsure. Thanks a million in advance.
Ok, so if you want to record the conversation of an active call, you will need to use the dx_mreciottdata, since it requires the TX timeslots from both directions. For instance recording a SIP call, you would get the timeslot of the IPM channel, and then timeslot of the dxxx device doing the plays of prompts. Or maybe you have an hairpinned IP call, so you get the timeslot of the other IPM channels. This was it records the conversation from both sides.
Otherwise, if you only perform regular record you are only getting one direction of the conversation. Not sure which you are looking for, but both can be done in the same application or, separate application using an available DXX device then will not be targeted for use by the app itself. So you need extra voice channels in order to do so depending on how many concurrent record you want to perform.
Jeff
Jeff, Thanks a million for the help. If I understand you correctly, as long as I keep the code in the main engine I run async, I won't need another dxxx resource. Is that correct? Here's what I'm doing, which seems straightforward, but of course doesn't work.
int start_rec_call(int ch)
{
int voxdev, ipmdev, recdev;
DX_IOTT iott = {0};
DV_TPT tpt = {0};
DX_XPB xpb = {0};
SC_TSINFO voxtsinfo, ipmtsinfo;
long recslots[1024], voxscts, ipmscts;
iott.io_type = IO_DEV | IO_EOT;
iott.io_offset = 0;
iott.io_length = -1;
if((iott.io_fhandle = open("/audio/tmp/recordcall.vox",O_RDWR|O_TRUNC|O_CREAT,0666)) < 0) {
dlog(ch,"Error opening call recording file");
return(-1);
}
dx_clrtpt(&tpt,1);
tpt.tp_type = IO_EOT;
tpt.tp_termno = DX_LCOFF;
tpt.tp_length = 1;
tpt.tp_flags = TF_LCOFF;
dlog(ch,"RECC:TPT setup");
xpb.wFileFormat = FILE_FORMAT_VOX;
xpb.wDataFormat = DATA_FORMAT_MULAW;
xpb.nSamplesPerSec = DRT_8KHZ;
xpb.wBitsPerSample = 8;
dlog(ch,"RECC:XPB setup");
voxtsinfo.sc_numts = 1;
voxtsinfo.sc_tsarrayp = &voxscts;
if(dx_getxmitslot(port[ch].voxfd,&voxtsinfo) == -1) {
dlog(ch,"Error on record call getsmitslot");
return(-1);
}
ipmtsinfo.sc_numts = 1;
ipmtsinfo.sc_tsarrayp = &ipmscts;
if(dx_getxmitslot(ipmdev,&ipmtsinfo) == -1) {
dlog(ch,"Error on record call getsmitslot");
return(-1);
}
recslots[1] = voxscts;
recslots[0] = ipmscts;
rectsinfo.sc_numts = 2;
rectsinfo.sc_tsarrayp = &recslots[0];
if(dx_mreciottdata(ipmdev,&iott, &tpt, &xpb, EV_ASYNC | TM_TONE, &rectsinfo) == -1) {
dlog(ch,"Error starting mreciottdata on vox dev");
dlog(ch,"Lasterr %d errmsg [%s]",ATDV_LASTERR(port[ch].voxfd),ATDV_ERRMSGP(port[ch].voxfd));
return(-1);
}
}