Something is wrong
Posted: Mon Jun 20, 2016 6:49 am
Hello all,
first things first. This forum has saved my life more than Fleming saved umanity
i run Vicidial for my call center and for partners.
The thing is this, i always used to work with vicibox 4.0.0(or 4.0.3, always preload iso-s) with updated svn for every instalation. asterisk 1.4,suse 12. old stuff, but goooood stuff.
never had problems. well sometime hdd breaks but i gues is nobodys fault.
than a took this wonderfull server. 2xquads (16threads) , 16ram. raid1 (2x125gb) SSD. i thought: "is time to move forward, so lets install vicibox 6.0.4. i was very happy about that. it has to keep alive something like 25 agents whith 6-9 adapt-tapered-ratio.
Here comes the nightmare. over 20 agents the server started to go crazy. at some point i was having only errors and warnings.
1) Unable to write to alert pipe on IAX2/127.0.0.1:41569-9628 (qlen = 77): Resource temporarily - a lot of them
2) exceptionally long voice queue length queuing to iax2 - a lot
3) timeout reached on 60 sec.
when i looked on sip show channels it gave me more calls than channels. i was forced to work a week like that, and at the first sunday BOOOOOM. installed back the old mighty vicibox 4.0.0 and everything is fine. now i have logged in 35 agents and averything works like a charm.
is the same machine, same ratio, same everything...
so my question is simple:
Why? Why in the same conditions vicibox 4 works much better than vicibox 6?
last thing, same isuee in other places. same problems in 3 other colleagues call centers.
Architecture: i686
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 2
NUMA node(s): 2
Vendor ID: GenuineIntel
CPU family: 6
Model: 44
Stepping: 2
CPU MHz: 1596.000
BogoMIPS: 4800.18
Virtualization: VT-x
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 12288K
NUMA node0 CPU(s): 0-3,8-11
NUMA node1 CPU(s): 4-7,12-15
(raid1 2x125gb SSD)
/dev/sda1:
Timing O_DIRECT disk reads: 874 MB in 3.00 seconds = 291.02 MB/sec
RAM: 4 x 4GB DDR3 ECC.
VERSION: 2.12-559a
BUILD: 160611-2230
SVN Version: 2548
first things first. This forum has saved my life more than Fleming saved umanity
i run Vicidial for my call center and for partners.
The thing is this, i always used to work with vicibox 4.0.0(or 4.0.3, always preload iso-s) with updated svn for every instalation. asterisk 1.4,suse 12. old stuff, but goooood stuff.
never had problems. well sometime hdd breaks but i gues is nobodys fault.
than a took this wonderfull server. 2xquads (16threads) , 16ram. raid1 (2x125gb) SSD. i thought: "is time to move forward, so lets install vicibox 6.0.4. i was very happy about that. it has to keep alive something like 25 agents whith 6-9 adapt-tapered-ratio.
Here comes the nightmare. over 20 agents the server started to go crazy. at some point i was having only errors and warnings.
1) Unable to write to alert pipe on IAX2/127.0.0.1:41569-9628 (qlen = 77): Resource temporarily - a lot of them
2) exceptionally long voice queue length queuing to iax2 - a lot
3) timeout reached on 60 sec.
when i looked on sip show channels it gave me more calls than channels. i was forced to work a week like that, and at the first sunday BOOOOOM. installed back the old mighty vicibox 4.0.0 and everything is fine. now i have logged in 35 agents and averything works like a charm.
is the same machine, same ratio, same everything...
so my question is simple:
Why? Why in the same conditions vicibox 4 works much better than vicibox 6?
last thing, same isuee in other places. same problems in 3 other colleagues call centers.
Architecture: i686
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 16
On-line CPU(s) list: 0-15
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 2
NUMA node(s): 2
Vendor ID: GenuineIntel
CPU family: 6
Model: 44
Stepping: 2
CPU MHz: 1596.000
BogoMIPS: 4800.18
Virtualization: VT-x
L1d cache: 32K
L1i cache: 32K
L2 cache: 256K
L3 cache: 12288K
NUMA node0 CPU(s): 0-3,8-11
NUMA node1 CPU(s): 4-7,12-15
(raid1 2x125gb SSD)
/dev/sda1:
Timing O_DIRECT disk reads: 874 MB in 3.00 seconds = 291.02 MB/sec
RAM: 4 x 4GB DDR3 ECC.
VERSION: 2.12-559a
BUILD: 160611-2230
SVN Version: 2548