[Ntop] starting ntop errors and warnings

[Spiros Papageorgiou]

Hi all,


I have an updated ntopng/pfring-zc installation and when I start ntopng
the following messages are produced:

Aug 30 18:05:06 server logger: ntopng start
Aug 30 18:05:08 server ntopng: [ViewInterface.cpp:50] ERROR: Internal
Error: NULL interface
[view:zc:***@0,zc:***@1,zc:***@2,zc:***@3][2]
Aug 30 18:05:08 server ntopng: [ViewInterface.cpp:50] ERROR: Internal
Error: NULL interface
[view:zc:***@0,zc:***@1,zc:***@2,zc:***@3][2]
Aug 30 18:05:08 server ntopng: [ViewInterface.cpp:50] ERROR: Internal
Error: NULL interface
[view:zc:***@0,zc:***@1,zc:***@2,zc:***@3][2]
Aug 30 18:05:08 server ntopng: [ViewInterface.cpp:50] ERROR: Internal
Error: NULL interface
[view:zc:***@0,zc:***@1,zc:***@2,zc:***@3][2]
Aug 30 18:05:08 server kernel: device ens160 entered promiscuous mode
Aug 30 18:05:09 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable
to set packet capture direction
Aug 30 18:05:11 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable
to set packet capture direction
Aug 30 18:05:11 server ntopng: Starting ntopng: Unable to start
ntopng[FAILED]
Aug 30 18:05:11 server systemd: Started Start/stop ntopng program.
Aug 30 18:05:12 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable
to set packet capture direction
Aug 30 18:05:14 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable
to set packet capture direction
Aug 30 18:05:14 server kernel: traps: ntopng[2575] general protection
ip:42df01 sp:7ff418f69b60 error:0 in ntopng[400000+264000]
Aug 30 18:05:15 server kernel: device ens160 left promiscuous mode
Aug 30 18:05:15 server systemd: ntopng.service: main process exited,
code=killed, status=11/SEGV

Why the view of all queues does not work? How should i write this line?

When I delete from ntopng.conf, the
"view:zc:***@0,zc:***@1,zc:***@2,zc:***@3" line, the I get
another bunch of warnings:

Aug 30 18:06:50 server systemd: Starting Start/stop ntopng program...
Aug 30 18:06:50 server logger: ntopng start
Aug 30 18:06:51 server kernel: device ens160 entered promiscuous mode
Aug 30 18:06:52 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable
to set packet capture direction
Aug 30 18:06:53 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable
to set packet capture direction
Aug 30 18:06:55 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable
to set packet capture direction
Aug 30 18:06:55 server ntopng: Starting ntopng: Unable to start
ntopng[FAILED]
Aug 30 18:06:55 server systemd: Started Start/stop ntopng program.
Aug 30 18:06:56 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable
to set packet capture direction
Aug 30 18:06:57 server ZC[2647]: error opening
/dev/hugepages/***@1-txrx: Permission denied
Aug 30 18:06:57 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error opening
/dev/hugepages/***@0-txrx: Permission denied
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error opening
/dev/hugepages/***@2-txrx: Permission denied
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error opening
/dev/hugepages/***@3-txrx: Permission denied
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:07:01 server systemd: Started Session 6 of user root.
Aug 30 18:07:01 server systemd: Starting Session 6 of user root.

The ntopng then, seems to be working but I would like to get rid of the
errors. What should I do to avoid the hugepages errors? I'm running ntop
as root (though I can see it drops privs).

My conf:

-G=/var/run/ntopng.pid
--data-dir=/mnt/data/ntopng
#--interface=zc:ens160
--interface=zc:***@0
--interface=zc:***@1
--interface=zc:***@2
--interface=zc:***@3
#--interface=view:zc:***@0,zc:***@1,zc:***@2,zc:***@3
--local-networks="xx.xx.0.0/16,xx.xx.cc.0/24,xx.xx.ff.0/22,xx.xx.vv.0/24"
--dns-mode=3
--disable-autologout
--max-num-flows=512000
--max-num-hosts=256000
--sticky-hosts=none
-g -1

# cat /etc/ntopng/ntopng.start
#

# cat /etc/pf_ring/hugepages.conf
node=0 hugepages=1024
# cat /etc/pf_ring/pf_ring.conf
transparent_mode=2

# cat /etc/pf_ring/pf_ring.start
# cat /etc/pf_ring/zc/ixgbe/ixgbe.conf
RSS=0,0,0,0
# cat /etc/pf_ring/zc/ixgbe/ixgbe.start
#



Regards,

Spiros

[Spiros Papageorgiou]

Hi Alfredo,

Here is the output:
[***@server ~]# cat /proc/mounts | grep huge
cgroup /sys/fs/cgroup/hugetlb cgroup
rw,nosuid,nodev,noexec,relatime,hugetlb 0 0
hugetlbfs /dev/hugepages hugetlbfs rw,relatime 0 0

[***@server ~]# cat /sys/devices/system/node/node*/meminfo | grep Huge
Node 0 AnonHugePages:   2488320 kB
Node 0 HugePages_Total:  2048
Node 0 HugePages_Free:   2048
Node 0 HugePages_Surp:      0
[***@server ~]#

Regards,
Sp

Hi Spiros
cat /proc/mounts / grep huge
cat /sys/devices/system/node/node*/meminfo | grep Huge
Alfredo

Post by Spiros Papageorgiou
Hi all,
Aug 30 18:05:06 server logger: ntopng start
Aug 30 18:05:08 server kernel: device ens160 entered promiscuous mode
Aug 30 18:05:09 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:11 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:11 server ntopng: Starting ntopng: Unable to start ntopng[FAILED]
Aug 30 18:05:11 server systemd: Started Start/stop ntopng program.
Aug 30 18:05:12 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:14 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:14 server kernel: traps: ntopng[2575] general protection ip:42df01 sp:7ff418f69b60 error:0 in ntopng[400000+264000]
Aug 30 18:05:15 server kernel: device ens160 left promiscuous mode
Aug 30 18:05:15 server systemd: ntopng.service: main process exited, code=killed, status=11/SEGV
Why the view of all queues does not work? How should i write this line?
Aug 30 18:06:50 server systemd: Starting Start/stop ntopng program...
Aug 30 18:06:50 server logger: ntopng start
Aug 30 18:06:51 server kernel: device ens160 entered promiscuous mode
Aug 30 18:06:52 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:53 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:55 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:55 server ntopng: Starting ntopng: Unable to start ntopng[FAILED]
Aug 30 18:06:55 server systemd: Started Start/stop ntopng program.
Aug 30 18:06:56 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:57 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:07:01 server systemd: Started Session 6 of user root.
Aug 30 18:07:01 server systemd: Starting Session 6 of user root.
The ntopng then, seems to be working but I would like to get rid of the errors. What should I do to avoid the hugepages errors? I'm running ntop as root (though I can see it drops privs).
-G=/var/run/ntopng.pid
--data-dir=/mnt/data/ntopng
#--interface=zc:ens160
--local-networks="xx.xx.0.0/16,xx.xx.cc.0/24,xx.xx.ff.0/22,xx.xx.vv.0/24"
--dns-mode=3
--disable-autologout
--max-num-flows=512000
--max-num-hosts=256000
--sticky-hosts=none
-g -1
# cat /etc/ntopng/ntopng.start
#
# cat /etc/pf_ring/hugepages.conf
node=0 hugepages=1024
# cat /etc/pf_ring/pf_ring.conf
transparent_mode=2
# cat /etc/pf_ring/pf_ring.start
# cat /etc/pf_ring/zc/ixgbe/ixgbe.conf
RSS=0,0,0,0
# cat /etc/pf_ring/zc/ixgbe/ixgbe.start
#
Regards,
Spiros
_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

[Spiros Papageorgiou]

Hi Luca,

I updated everything and the error messages seem to be gone. I can see
that huge pages have been allocated:

# cat /sys/devices/system/node/node*/meminfo | grep Huge F10Quit
Node 0 AnonHugePages:    100352 kB
Node 0 HugePages_Total:  2048
Node 0 HugePages_Free:   1852
Node 0 HugePages_Surp:      0

Thanx!

Two more problems:
- What about the "view:zc:***@0,zc:***@1,zc:***@2,zc:***@3"
which is not working? Is the syntax wrong or something?
- I get "02/Sep/2017 07:34:50 [Lua.cpp:5629]
[@/usr/share/ntopng/scripts/callbacks/5min.lua]:[161] ERROR: Cannot
complete local hosts RRD dump in 5 minutes. Please check your RRD
configuration.". Do I need a stronger server? I have a 4core VM  and the
disk (SAN) seems not to be stressed that much. The CPU though is at
about 70%.
- The same goes when I export to ES. The ES is at about 30% CPU
utilization but ntopng reports massive flow drops.

Regards,
Spiros

Post by Spiros Papageorgiou
Spiros
can you please update ntopng and report?
Thanks Luca

Post by Spiros Papageorgiou
Hi Alfredo,
cgroup /sys/fs/cgroup/hugetlb cgroup
rw,nosuid,nodev,noexec,relatime,hugetlb 0 0
hugetlbfs /dev/hugepages hugetlbfs rw,relatime 0 0
Node 0 AnonHugePages:   2488320 kB
Node 0 HugePages_Total:  2048
Node 0 HugePages_Free:   2048
Node 0 HugePages_Surp:      0
Regards,
Sp

Hi Spiros
cat /proc/mounts / grep huge
cat /sys/devices/system/node/node*/meminfo | grep Huge
Alfredo

Post by Spiros Papageorgiou
Hi all,
Aug 30 18:05:06 server logger: ntopng start
Aug 30 18:05:08 server kernel: device ens160 entered promiscuous mode
Aug 30 18:05:09 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:11 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:11 server ntopng: Starting ntopng: Unable to start ntopng[FAILED]
Aug 30 18:05:11 server systemd: Started Start/stop ntopng program.
Aug 30 18:05:12 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:14 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:14 server kernel: traps: ntopng[2575] general protection ip:42df01 sp:7ff418f69b60 error:0 in ntopng[400000+264000]
Aug 30 18:05:15 server kernel: device ens160 left promiscuous mode
Aug 30 18:05:15 server systemd: ntopng.service: main process exited, code=killed, status=11/SEGV
Why the view of all queues does not work? How should i write this line?
Aug 30 18:06:50 server systemd: Starting Start/stop ntopng program...
Aug 30 18:06:50 server logger: ntopng start
Aug 30 18:06:51 server kernel: device ens160 entered promiscuous mode
Aug 30 18:06:52 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:53 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:55 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:55 server ntopng: Starting ntopng: Unable to start ntopng[FAILED]
Aug 30 18:06:55 server systemd: Started Start/stop ntopng program.
Aug 30 18:06:56 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:57 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:07:01 server systemd: Started Session 6 of user root.
Aug 30 18:07:01 server systemd: Starting Session 6 of user root.
The ntopng then, seems to be working but I would like to get rid of the errors. What should I do to avoid the hugepages errors? I'm running ntop as root (though I can see it drops privs).
-G=/var/run/ntopng.pid
--data-dir=/mnt/data/ntopng
#--interface=zc:ens160
--local-networks="xx.xx.0.0/16,xx.xx.cc <http://xx.xx.cc>.0/24,xx.xx.ff.0/22,xx.xx.vv.0/24"
--dns-mode=3
--disable-autologout
--max-num-flows=512000
--max-num-hosts=256000
--sticky-hosts=none
-g -1
# cat /etc/ntopng/ntopng.start
#
# cat /etc/pf_ring/hugepages.conf
node=0 hugepages=1024
# cat /etc/pf_ring/pf_ring.conf
transparent_mode=2
# cat /etc/pf_ring/pf_ring.start
# cat /etc/pf_ring/zc/ixgbe/ixgbe.conf
RSS=0,0,0,0
# cat /etc/pf_ring/zc/ixgbe/ixgbe.start
#
Regards,
Spiros
_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

[Spiros Papageorgiou]

yeap, ok.

Thanx!
Sp

Spiros,
can you please file an issue on github s we can track this problem?
Please open one ticket per issue
thanks Luca

Post by Spiros Papageorgiou
Hi Luca,
I updated everything and the error messages seem to be gone. I can
# cat /sys/devices/system/node/node*/meminfo | grep Huge F10Quit
Node 0 AnonHugePages:    100352 kB
Node 0 HugePages_Total:  2048
Node 0 HugePages_Free:   1852
Node 0 HugePages_Surp:      0
Thanx!
- What about the
working? Is the syntax wrong or something?
- I get "02/Sep/2017 07:34:50 [Lua.cpp:5629]
complete local hosts RRD dump in 5 minutes. Please check your RRD
configuration.". Do I need a stronger server? I have a 4core VM  and
the disk (SAN) seems not to be stressed that much. The CPU though is
at about 70%.
- The same goes when I export to ES. The ES is at about 30% CPU
utilization but ntopng reports massive flow drops.
Regards,
Spiros

Post by Spiros Papageorgiou
Spiros
can you please update ntopng and report?
Thanks Luca

Post by Spiros Papageorgiou
Hi Alfredo,
cgroup /sys/fs/cgroup/hugetlb cgroup
rw,nosuid,nodev,noexec,relatime,hugetlb 0 0
hugetlbfs /dev/hugepages hugetlbfs rw,relatime 0 0
Node 0 AnonHugePages: 2488320 kB
Node 0 HugePages_Total: 2048
Node 0 HugePages_Free: 2048
Node 0 HugePages_Surp:      0
Regards,
Sp

Hi Spiros
cat /proc/mounts / grep huge
cat /sys/devices/system/node/node*/meminfo | grep Huge
Alfredo

Post by Spiros Papageorgiou
Hi all,
Aug 30 18:05:06 server logger: ntopng start
Aug 30 18:05:08 server kernel: device ens160 entered promiscuous mode
Aug 30 18:05:09 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:11 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:11 server ntopng: Starting ntopng: Unable to start ntopng[FAILED]
Aug 30 18:05:11 server systemd: Started Start/stop ntopng program.
Aug 30 18:05:12 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:14 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:14 server kernel: traps: ntopng[2575] general protection ip:42df01 sp:7ff418f69b60 error:0 in ntopng[400000+264000]
Aug 30 18:05:15 server kernel: device ens160 left promiscuous mode
Aug 30 18:05:15 server systemd: ntopng.service: main process exited, code=killed, status=11/SEGV
Why the view of all queues does not work? How should i write this line?
Aug 30 18:06:50 server systemd: Starting Start/stop ntopng program...
Aug 30 18:06:50 server logger: ntopng start
Aug 30 18:06:51 server kernel: device ens160 entered promiscuous mode
Aug 30 18:06:52 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:53 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:55 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:55 server ntopng: Starting ntopng: Unable to start ntopng[FAILED]
Aug 30 18:06:55 server systemd: Started Start/stop ntopng program.
Aug 30 18:06:56 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:57 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:07:01 server systemd: Started Session 6 of user root.
Aug 30 18:07:01 server systemd: Starting Session 6 of user root.
The ntopng then, seems to be working but I would like to get rid of the errors. What should I do to avoid the hugepages errors? I'm running ntop as root (though I can see it drops privs).
-G=/var/run/ntopng.pid
--data-dir=/mnt/data/ntopng
#--interface=zc:ens160
--local-networks="xx.xx.0.0/16,xx.xx.cc <http://xx.xx.cc/>.0/24,xx.xx.ff.0/22,xx.xx.vv.0/24"
--dns-mode=3
--disable-autologout
--max-num-flows=512000
--max-num-hosts=256000
--sticky-hosts=none
-g -1
# cat /etc/ntopng/ntopng.start
#
# cat /etc/pf_ring/hugepages.conf
node=0 hugepages=1024
# cat /etc/pf_ring/pf_ring.conf
transparent_mode=2
# cat /etc/pf_ring/pf_ring.start
# cat /etc/pf_ring/zc/ixgbe/ixgbe.conf
RSS=0,0,0,0
# cat /etc/pf_ring/zc/ixgbe/ixgbe.start
#
Regards,
Spiros
_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

[Spiros Papageorgiou]

Hi all,

I tried to export flows from ntopng to ELK and ntopng reports:
Exported Flows 95,984 [568.57 Flows/s] Dropped Flows 871,640 [908.11%]

My ELK host has low CPU usage (about 20%) and the storage seems to have
about 1K-2K IOPs.

I would appreciate if anyone has information regarding an ELK cluster
that can handle about 10K Flows/s and be able to respond to queries. I
need the info to dimension my needs.

Sorry for a question that is not direclty related with ntopng, but I
need info about the specific load that ntopng puts on the ELK cluster.

Thank you,
Spiros

[Spiros Papageorgiou]

Spiros,

Post by Spiros Papageorgiou
Hi Luca,
I updated everything and the error messages seem to be gone. I can
# cat /sys/devices/system/node/node*/meminfo | grep Huge F10Quit
Node 0 AnonHugePages:    100352 kB
Node 0 HugePages_Total:  2048
Node 0 HugePages_Free:   1852
Node 0 HugePages_Surp:      0
Thanx!
- What about the
working? Is the syntax wrong or something?

What do you mean with 'is not working'? So you get empty pages?
Please, report the issue using our GitHub issue tracker.

Post by Spiros Papageorgiou
- I get "02/Sep/2017 07:34:50 [Lua.cpp:5629]
complete local hosts RRD dump in 5 minutes. Please check your RRD
configuration.". Do I need a stronger server? I have a 4core VM  and
the disk (SAN) seems not to be stressed that much. The CPU though is
at about 70%.

This means that you have too many local hosts that there's not enough
time to update their statistics in 5 minutes. How many local hosts do
you have? Disabling data retention of L7 timeseries may alleviate the
issue.
The bottleneck here are the several multiple RRD updates made for
every local host.

Ntopng reports about 10K local hosts and 50K remote.
How can i disable L7 timeseries data retention? Can I keep stats for
only local hosts? What kind of control do i have on that aspect?
Is it possible to use somekind of caching, like rrdcache?

Post by Spiros Papageorgiou
- The same goes when I export to ES. The ES is at about 30% CPU
utilization but ntopng reports massive flow drops.

ElasticSearch is not fast enough to inject flows at the required
speed. Hence ntopng flows export queue grows and at some point it
overflows resulting in flows drops. Other people experienced this in
the past. You must configure your ES deployment to make sure it has
enough intake.

Post by Spiros Papageorgiou
Regards,
Spiros

Post by Spiros Papageorgiou
Spiros
can you please update ntopng and report?
Thanks Luca

Post by Spiros Papageorgiou
Hi Alfredo,
cgroup /sys/fs/cgroup/hugetlb cgroup
rw,nosuid,nodev,noexec,relatime,hugetlb 0 0
hugetlbfs /dev/hugepages hugetlbfs rw,relatime 0 0
Node 0 AnonHugePages: 2488320 kB
Node 0 HugePages_Total: 2048
Node 0 HugePages_Free: 2048
Node 0 HugePages_Surp:      0
Regards,
Sp

Hi Spiros
cat /proc/mounts / grep huge
cat /sys/devices/system/node/node*/meminfo | grep Huge
Alfredo

Post by Spiros Papageorgiou
Hi all,
Aug 30 18:05:06 server logger: ntopng start
Aug 30 18:05:08 server kernel: device ens160 entered promiscuous mode
Aug 30 18:05:09 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:11 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:11 server ntopng: Starting ntopng: Unable to start ntopng[FAILED]
Aug 30 18:05:11 server systemd: Started Start/stop ntopng program.
Aug 30 18:05:12 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:14 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:05:14 server kernel: traps: ntopng[2575] general protection ip:42df01 sp:7ff418f69b60 error:0 in ntopng[400000+264000]
Aug 30 18:05:15 server kernel: device ens160 left promiscuous mode
Aug 30 18:05:15 server systemd: ntopng.service: main process exited, code=killed, status=11/SEGV
Why the view of all queues does not work? How should i write this line?
Aug 30 18:06:50 server systemd: Starting Start/stop ntopng program...
Aug 30 18:06:50 server logger: ntopng start
Aug 30 18:06:51 server kernel: device ens160 entered promiscuous mode
Aug 30 18:06:52 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:53 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:55 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:55 server ntopng: Starting ntopng: Unable to start ntopng[FAILED]
Aug 30 18:06:55 server systemd: Started Start/stop ntopng program.
Aug 30 18:06:56 server ntopng: [PF_RINGInterface.cpp:72] WARNING: Unable to set packet capture direction
Aug 30 18:06:57 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:06:58 server ZC[2647]: error mmap'ing 49 hugepages of 2048 KB
Aug 30 18:07:01 server systemd: Started Session 6 of user root.
Aug 30 18:07:01 server systemd: Starting Session 6 of user root.
The ntopng then, seems to be working but I would like to get rid of the errors. What should I do to avoid the hugepages errors? I'm running ntop as root (though I can see it drops privs).
-G=/var/run/ntopng.pid
--data-dir=/mnt/data/ntopng
#--interface=zc:ens160
--local-networks="xx.xx.0.0/16,xx.xx.cc <http://xx.xx.cc/>.0/24,xx.xx.ff.0/22,xx.xx.vv.0/24"
--dns-mode=3
--disable-autologout
--max-num-flows=512000
--max-num-hosts=256000
--sticky-hosts=none
-g -1
# cat /etc/ntopng/ntopng.start
#
# cat /etc/pf_ring/hugepages.conf
node=0 hugepages=1024
# cat /etc/pf_ring/pf_ring.conf
transparent_mode=2
# cat /etc/pf_ring/pf_ring.start
# cat /etc/pf_ring/zc/ixgbe/ixgbe.conf
RSS=0,0,0,0
# cat /etc/pf_ring/zc/ixgbe/ixgbe.start
#
Regards,
Spiros
_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop

_______________________________________________
Ntop mailing list
http://listgateway.unipi.it/mailman/listinfo/ntop