Hello!
I'm switching to Debian 8 from Debian 7 and found very interesting bug
with your load_driver.sh (ixgbe).
./load_driver.sh
rmmod: ERROR: Module ixgbe is not currently loaded
Warning: 0 hugepages available, 1024 requested
irqbalance: no process found
Configuring eth6
no rx vectors found on eth6
no tx vectors found on eth6
eth6 mask=1 for /proc/irq/112/smp_affinity
Configuring eth7
no rx vectors found on eth7
no tx vectors found on eth7
eth7 mask=1 for /proc/irq/114/smp_affinity
Configuring eth4
no rx vectors found on eth4
no tx vectors found on eth4
eth4 mask=1 for /proc/irq/108/smp_affinity
Configuring eth5
no rx vectors found on eth5
no tx vectors found on eth5
eth5 mask=1 for /proc/irq/110/smp_affinity
As you can see load_driver.sh tried to allocate 1024 huge pages but
failed (maybe you should fail in this case?).
This issue related with allocated by default in Debian 8 hugetlb buffers:
AnonHugePages: 0 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 2048 kB
DirectMap4k: 74240 kB
DirectMap2M: 50255872 kB
But there are no allocated pages:
grep HugePages_Total /sys/devices/system/node/node0/meminfo
I suggest following fix for load_driver.sh:
if [ `cat /proc/mounts | grep hugetlbfs | wc -l` -eq 0 ]; then
sync && echo 3 > /proc/sys/vm/drop_caches
echo $HUGEPAGES > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
mkdir /mnt/huge
mount -t hugetlbfs nodev /mnt/huge
else
echo $HUGEPAGES > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
fi
After this fix everything loaded correctly:
./load_driver.sh
rmmod: ERROR: Module ixgbe is not currently loaded
rmmod: ERROR: Module pf_ring is not currently loaded
Configuring eth6
no rx vectors found on eth6
no tx vectors found on eth6
eth6 mask=1 for /proc/irq/112/smp_affinity
Configuring eth7
no rx vectors found on eth7
no tx vectors found on eth7
eth7 mask=1 for /proc/irq/114/smp_affinity
Configuring eth4
no rx vectors found on eth4
no tx vectors found on eth4
eth4 mask=1 for /proc/irq/108/smp_affinity
Configuring eth5
no rx vectors found on eth5
no tx vectors found on eth5
eth5 mask=1 for /proc/irq/110/smp_affinity
#########################################################################
# ERROR: You do not seem to have a valid PF_RING ZC license
6.0.3.150330 for eth4 [Intel 10 Gbit ixgbe 82599-based]
# ERROR: Please get one at http://shop.ntop.org/.
#########################################################################
# We're now working in demo mode with packet capture and
# transmission limited to 5 minutes
#########################################################################
libnuma: Warning: node 6 not allowed
numa_sched_setaffinity_v2_int() failed; abort
: Invalid argument
set_mempolicy: Invalid argument
#########################################################################
# ERROR: You do not seem to have a valid PF_RING ZC license
6.0.3.150330 for eth4 [Intel 10 Gbit ixgbe 82599-based]
# ERROR: Please get one at http://shop.ntop.org/.
#########################################################################
#########################################################################
# ERROR: You do not seem to have a valid PF_RING ZC license
6.0.3.150330 for eth4 [Intel 10 Gbit ixgbe 82599-based]
# ERROR: Please get one at http://shop.ntop.org/.
#########################################################################
^CLeaving...
=========================
Absolute Stats: 0 pkts - 0 bytes
=========================
--
Sincerely yours, Pavel Odintsov
_______________________________________________
Ntop-misc mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop-misc
