Syswatch Issues with ping fail

Offline

Issue

Syswatch Issues with ping fail

Resolved

0 votes

I keep getting ping fails that will cause my primary to switch to backup. How can I change the ping server list or patch this issue?

Mon Feb 15 19:38:26 2016 info: system - heartbeat...
Mon Feb 15 19:40:28 2016 info: enp2s0 - ping check on server #1 failed - 8.8.8.8
Mon Feb 15 19:40:33 2016 info: enp2s0 - ping check on server #2 failed - 54.152.208.245
Mon Feb 15 19:40:33 2016 warn: enp2s0 - connection warning
Mon Feb 15 19:40:44 2016 info: enp2s0 - ping check on server #1 passed - 8.8.8.8

Tue Feb 16 10:12:42 2016 info: system - heartbeat...
Tue Feb 16 10:15:45 2016 info: enp1s0 - ping check on server #1 failed - 8.8.8.8
Tue Feb 16 10:15:50 2016 info: enp1s0 - ping check on server #2 failed - 54.152.208.245
Tue Feb 16 10:15:50 2016 warn: enp1s0 - connection warning
Tue Feb 16 10:16:02 2016 info: enp1s0 - ping check on gateway failed - 75.148.200.246
Tue Feb 16 10:16:04 2016 info: enp1s0 - ping check on server #1 failed - 8.8.8.8
Tue Feb 16 10:16:09 2016 info: enp1s0 - ping check on server #2 failed - 54.152.208.245
Tue Feb 16 10:16:09 2016 warn: enp1s0 - connection warning
Tue Feb 16 10:16:21 2016 info: enp1s0 - ping check on server #1 failed - 8.8.8.8
Tue Feb 16 10:16:26 2016 info: enp1s0 - ping check on server #2 failed - 54.152.208.245
Tue Feb 16 10:16:26 2016 warn: enp1s0 - upstream connection problems with your ISP
Tue Feb 16 10:16:26 2016 info: system - changing active WAN list - enp2s0 (was enp1s0 enp2s0)
Tue Feb 16 10:16:26 2016 info: system - current WANs in use - enp2s0
Tue Feb 16 10:16:26 2016 warn: system - default route is not active... removing - enp1s0
Tue Feb 16 10:16:26 2016 info: system - restarting firewall
Tue Feb 16 10:16:30 2016 info: system - default route vanished - using enp2s0
Tue Feb 16 10:16:30 2016 info: system - setting default route to 10.1.10.1
Tue Feb 16 10:16:30 2016 info: system - restarting firewall
Tue Feb 16 10:16:34 2016 info: system - restarting firewall
Tue Feb 16 10:16:38 2016 info: system - updating intrusion prevention whitelist
Tue Feb 16 10:16:38 2016 info: system - adding ping server 54.152.208.245
Tue Feb 16 10:16:38 2016 info: system - adding ping server 8.8.8.8
Tue Feb 16 10:16:38 2016 info: system - adding DNS server 75.75.75.75
Tue Feb 16 10:16:38 2016 info: system - adding DNS server 75.75.76.76
Tue Feb 16 10:16:50 2016 info: enp1s0 - ping check on server #1 failed - 8.8.8.8
Tue Feb 16 10:16:55 2016 info: enp1s0 - ping check on server #2 failed - 54.152.208.245
Tue Feb 16 10:16:55 2016 warn: enp1s0 - upstream connection problems with your ISP
Tue Feb 16 10:16:57 2016 info: enp1s0 - waiting for static IP reconnect
Tue Feb 16 10:17:07 2016 info: enp1s0 - ping check on server #1 passed - 8.8.8.8
Tue Feb 16 10:17:07 2016 info: system - changing active WAN list - enp1s0 enp2s0 (was enp2s0)
Tue Feb 16 10:17:07 2016 info: system - current WANs in use - enp1s0
Tue Feb 16 10:17:07 2016 warn: system - default route is not active... removing - enp2s0
Tue Feb 16 10:17:07 2016 info: system - restarting firewall
Tue Feb 16 10:17:11 2016 info: system - default route vanished - using enp1s0
Tue Feb 16 10:17:11 2016 info: system - setting default route to 75.148.200.246
Tue Feb 16 10:17:11 2016 info: system - restarting firewall
Tue Feb 16 10:17:15 2016 info: system - restarting firewall
Tue Feb 16 10:17:18 2016 info: system - updating intrusion prevention whitelist
Tue Feb 16 10:17:18 2016 info: system - adding ping server 54.152.208.245
Tue Feb 16 10:17:18 2016 info: system - adding ping server 8.8.8.8
Tue Feb 16 10:17:18 2016 info: system - adding DNS server 75.75.75.75
Tue Feb 16 10:17:18 2016 info: system - adding DNS server 75.75.76.76
Tue Feb 16 10:17:58 2016 info: system - heartbeat...
Tue Feb 16 10:23:00 2016 info: system - heartbeat...

lspci -k | grep Eth -A 3
01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 P CI Express Gigabit Ethernet Controller (rev 06)
Subsystem: TP-LINK Technologies Co., Ltd. TG-3468 Gigabit PCI Express Ne twork Adapter
Kernel driver in use: r8169
02:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network Connection
Subsystem: Super Micro Computer Inc X7SPA-H/X7SPA-HF Motherboard
Kernel driver in use: e1000e
03:00.0 Ethernet controller: Intel Corporation 82574L Gigabit Network Connection
Subsystem: Super Micro Computer Inc X7SPA-H/X7SPA-HF Motherboard
Kernel driver in use: e1000e

Multi-WAN

In Multi-WAN

Tuesday, February 16 2016, 04:53 PM

Share this post:

Responses (24)

Accepted Answer

David Smith

Offline
Thursday, May 25 2017, 05:18 PM - #Permalink
Resolved

0 votes

Nick Howitt wrote:

My ISP removed its free webspace so the files are now here. I don't think they will necessarily help. You only need the kmod-r8169 driver if you also need the kmod-r8168 driver, and you don't as you don't have an r8168 compatible card.

Thank you very much Nick - I'll give it a whirl and worst case I'll replace the NICs !
The reply is currently minimized Show
Accepted Answer
Nick Howitt

Offline
Thursday, May 25 2017, 04:31 PM - #Permalink
Resolved

0 votes

My ISP removed its free webspace so the files are now here. I don't think they will necessarily help. You only need the kmod-r8169 driver if you also need the kmod-r8168 driver, and you don't as you don't have an r8168 compatible card.
The reply is currently minimized Show
Accepted Answer

David Smith

Offline
Thursday, May 25 2017, 12:07 PM - #Permalink
Resolved

0 votes

Nick Howitt wrote:

Hi Michael,
Your reply will get through when the mods approve it. I think it may be the last one requiring approval.

For the RTL NIC, please install this driver. For the 82574L please install this one. To install do a "yum localinstall --nogpgcheck ..." or "rpm -ivh ...". Reboot once they are installed.

In case you are wondering what the drivers are, they are the kmod drivers supplied by ElRepo which I've compiled for ClearOS. They are based on the original manufacturers' drivers but changed so they should not need recompiling every time the kernel goes through a minor update.

Hi Nick

I chanced on this after days of trying to sort out an issue where multiwan keeps dropping a connection - finally I looked here and the inability to ping 8.8.8.8 pointed me in the right direction .... but the link to your download is gone - do you still have drivers available for the RTL8169?

My system details :

# lspci -k | grep Eth -A 3 00:19.0 Ethernet controller: Intel Corporation 82578DC Gigabit Network Connection (rev 06) Subsystem: Intel Corporation Device 0037 Kernel driver in use: e1000e Kernel modules: e1000e -- 03:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8169 PCI Gigabit Ethernet Controller (rev 10) Subsystem: Realtek Semiconductor Co., Ltd. RTL8169/8110 Family PCI Gigabit Ethernet NIC Kernel driver in use: r8169 Kernel modules: r8169 03:01.0 Ethernet controller: VIA Technologies, Inc. VT6105/VT6106S [Rhine-III] (rev 8b) Subsystem: D-Link System Inc DFE-520TX Fast Ethernet PCI Adapter Kernel driver in use: via-rhine Kernel modules: via-rhine 03:02.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8169 PCI Gigabit Ethernet Controller (rev 10) Subsystem: Realtek Semiconductor Co., Ltd. RTL8169/8110 Family PCI Gigabit Ethernet NIC Kernel driver in use: r8169 Kernel modules: r8169

# uname -r 2.6.32-642.15.1.v6.x86_64

Many thanks!
The reply is currently minimized Show
Accepted Answer
Tony Ellis

Offline
Wednesday, February 15 2017, 02:50 AM - #Permalink
Resolved

0 votes

Would have expected /etc/sysconfig/network-scripts/ifcfg-ens33 and ..../ifcfg-ens37 to be the same except for DHCP_HOSTNAME. Any reason for the difference? Not sure it would make a difference but would be inclined to make ens37 the same as ens33 expect for that hostname...

Also a functional check for ens37 should be undertaken. If you turn off Multi-Wan and just configure ens37 as only external interface - does it work? You indicated earlier both interfaces are for the same ISP. Can you check when you do this test for ens37 that you get the same DNS addresses as that which you had when ens33 was in use...

EDIT:
Your externals are private addresses - so cannot be the real ISPs assigned addresses surely - what is assigning these? I forget now... Any possibility of changing to static for a test? If you do remove "GATEWAYDEV="ens37"" from /etc/sysconfig/network.
The reply is currently minimized Show

Accepted Answer

Cumhur ata

Offline

Tuesday, February 14 2017, 05:04 PM - #Permalink

Resolved

0 votes

Hi,

I am sorry for late answer.



[root@gateway ~]# cat /etc/sysconfig/network-scripts/ifcfg-ens33

DEVICE=ens33

TYPE="Ethernet"

ONBOOT="yes"

USERCTL="no"

BOOTPROTO="dhcp"

PEERDNS="yes"

DHCP_HOSTNAME="192.168.1.1"



[root@gateway ~]# cat /etc/sysconfig/network-scripts/ifcfg-ens37

DEVICE=ens37

TYPE="Ethernet"

ONBOOT="yes"

USERCTL="no"

BOOTPROTO="dhcp"

PEERDNS="no"



[root@gateway ~]# cat /etc/sysconfig/network

NOZEROCONF="yes"

GATEWAYDEV="ens37"

[root@gateway ~]#

The reply is currently minimized Show

Accepted Answer
Tony Ellis

Offline
Sunday, February 05 2017, 01:13 AM - #Permalink
Resolved

0 votes

Thanks for using code tags... you missed one of my requests... "ip address"

The only difference between yours and mine is this in mine... I don't use bandwidth control... but that shouldn't make a difference.

# ISP Maximum Speeds ENP0S16F4U1_MAX_DOWNSTREAM=0 ENP0S16F4U1_MAX_UPSTREAM=0 ENP0S18_MAX_DOWNSTREAM=0 ENP0S18_MAX_UPSTREAM=0 ENP0S9_MAX_DOWNSTREAM=0 ENP0S9_MAX_UPSTREAM=0

Now a further request please - contents of /etc/sysconfig/network-scripts/ifcfg-ens33 and corresponding ens37 - and /etc/sysconfig/network

Sorry for all these requests - but if all these conf files are correct - then I can only assume it has something to do with VM and virtual network interfaces - something I know nothing about...
The reply is currently minimized Show

Accepted Answer

Cumhur ata

Offline

Saturday, February 04 2017, 09:27 PM - #Permalink

Resolved

0 votes

here is my networkd.conf



[root@gateway ~]# cat /etc/clearos/network.conf 

# Network mode

MODE="gateway"



# Network interface roles

EXTIF="ens33 ens37"

LANIF="ens35"

DMZIF=""

HOTIF=""



# Domain and Internet Hostname

DEFAULT_DOMAIN="aysis.local"

INTERNET_HOSTNAME="gateway.aysis.local"



# Extra LANS

EXTRALANS=""



# ISP Maximum Speeds

ENS37_MAX_DOWNSTREAM=90000

ENS37_MAX_UPSTREAM=24000

ENS36_MAX_DOWNSTREAM=0

ENS36_MAX_UPSTREAM=0

ENS35_MAX_DOWNSTREAM=0

ENS35_MAX_UPSTREAM=0

ENS32_MAX_DOWNSTREAM=0

ENS32_MAX_UPSTREAM=0

ENS33_MAX_DOWNSTREAM=47200

ENS33_MAX_UPSTREAM=3740

and then multiwan.conf



[root@gateway ~]# cat /etc/clearos/multiwan.conf 

MULTIPATH="on"

MULTIPATH_WEIGHTS="ens37|50 ens33|75"

EXTIF_BACKUP=""

EXTIF_STANDBY=""

[root@gateway ~]#

The reply is currently minimized Show

Accepted Answer
Tony Ellis

Offline
Saturday, February 04 2017, 01:52 PM - #Permalink
Resolved

0 votes

You would not expect a route to normally be associated with a LAN connection, hence ip route doesn't list it...
Can you please show the full contents of "/etc/clearos/network.conf" and "/etc/clearos/multiwan.conf"
Also the output of "ip address"

and *please* use code tags - it makes it so much easier to read...

here's my output from the command you used... definitely more here...

[root@alex ~]# ip -4 route show table all | grep "default" | grep -v "prohibit" && ip -4 route show table all | grep nexthop default via 192.168.2.35 dev enp0s18 table 100 default via 192.168.6.1 dev enp0s9f0 table 101 default via 192.168.6.1 dev enp0s9f0 table 200 proto static src 192.168.6.28 default via 192.168.2.35 dev enp0s18 table 201 proto static src 192.168.2.28 default table 250 proto static default via 192.168.6.1 dev enp0s9f0 nexthop via 192.168.2.35 dev enp0s18 weight 10 nexthop via 192.168.6.1 dev enp0s9f0 weight 75 [root@alex ~]#

Past midnight here in Oz - so will look in the morning for any update... hopefully some-one else will respond in the meantime...
The reply is currently minimized Show
Accepted Answer

Cumhur ata

Offline
Saturday, February 04 2017, 12:14 PM - #Permalink
Resolved

0 votes

Here are my ip route table how it looks like. I think there is no nexthub. can anyone see if there is anything wrong with the settings below?
ens33 and ens37 are WAN and there is also only one LAN connection which is not shown below Its ip network 192.168.2.0/255.255.255.0)

[root@gateway ~]# ip -4 route show table all | grep "default" | grep -v "prohibit" && ip -4 route show table all | grep nexthop
default via 192.168.1.1 dev ens33 table 100
default via 192.168.3.1 dev ens37 table 101
default via 192.168.1.1 dev ens33 table 200 proto static src 192.168.1.22
default via 192.168.3.1 dev ens37 table 201 proto static src 192.168.3.253
default via 192.168.1.1 dev ens33 table 250
default via 192.168.1.1 dev ens33
[root@gateway ~]#
The reply is currently minimized Show
Accepted Answer

Steve Thompson

Offline
Sunday, January 29 2017, 04:33 PM - #Permalink
Resolved

0 votes

Tony Ellis wrote:

You definitely will get flaky behaviour with the incorrect driver. Difficult to know whether it will completely fix your problem, but would sure help... You can change the two ping addresses by editing the file "/etc/syswatch" - see the bottom few lines. An example :-

..... ..... # Ping servers # - The syswatch daemon by default will just ping the next hop on the Internet # and another device. You can override this with a comma separated list of # 2 servers. #ping_servers=111.222.333.444,555.666.777.888 ping_servers=8.8.8.8,8.8.4.4

Sorry for such a late reply (the results of moving, updating Linux releases, [re]wiring the new house w/ CAT6, problems with my paying job...]. But your answer hit the nail on the head. I've been dealing with other issues and I think I have them all fixed. I must say that I did have to install the correct drivers for the ethernet chips.

Also, I am going to give information in this post that was given prior (almost a year ago) to keep one from having to jump back and forth between my posts.

I'm currently running 6.7-1 w/ kernel-2.6.32-642.11.1.v6.x86_64 and all the associated update (auto-updating works ).

With that said, and for others having some issues -- I run two ISPs, one is fiber optic (eth0) and the other is cable modem based (eth3). But I am running a single LAN for my network.

Coming up to this point, after fixing issues and changing heartbeat IP addresses, I get ping failures at random points. Generally I get ping failures for 8.8.8.8 from the cable modem ISP (was an RR.COM provider, specifically BrightHouse Networks, now Spectrum) but not the fiber optic based provider. The second ping address (I forgot the IP address, but it is a publicly available, but private DNS service I use) always* responds.

Why did I change the IP addresses for heartbeat? Well, the default is to ping the first connection point, which is typically the address providing the IP address to that ethernet adapter. With the cable modems, they generally do IPv4 Class C to the customer. So the idea is ping the cable modem. Well, that seems to fail quite frequently in my case. So I started using a Google DNS IP, which does respond to pings. The same problem was happening with the fiber optic system, so I changed it to do the same address. This actually gives one a way of knowing if Google has problems if you can't ping it from either ISP. Second ping address is either for a ClearOS address or a private DNS

* When either ISP has a serious problem, neither ping works, and I get failover to the other ISP for all traffic. Or, in the case of someone with a backhoe just outside of our neighborhood, they cut both cable and fiber optic... The joys of infrastructure "repairs" here in our town.

Now, as a test of my system, I pulled the cable modem out of the system, hot. Spectrum is being a problem child and so I am looking for another ISP (I am blessed with having multiple providers where I am). This caused a failover to the one ISP, but syswatch is now filling up with complaints about not being able to connect to the ISP that was on eth3. However, a scan of my logs for eth0 having a problem shows NOTHING (typical for Metronet here where I am -- they just run, if people don't cut their fiber cables ;-) ).

Again, I did appreciate the answer and I effected it as soon as I could with the result being my system (from the Gateway/Firewall box) has been very stable.
The reply is currently minimized Show
Accepted Answer
Tony Ellis

Offline
Sunday, January 29 2017, 03:58 PM - #Permalink
Resolved

0 votes

I don't use /know VMs - so cannot help in that area...

However, the net-tools package is deprecated (and netsat is one of the tools within this package), and in the case of multi-wan, netstat doesn't show all the routes... The tool to use is "ip route"...
Here's an example from a ClearOS 7.3 system - I have used grep to reduce the output to the essentials. See the "man" pages for all the gory details...

[root@alex ~]# ip -4 route show table all | grep "default" | grep -v "prohibit" && ip -4 route show table all | grep nexthop default via 192.168.2.35 dev enp0s18 table 100 default via 192.168.6.1 dev enp0s9f0 table 101 default via 192.168.6.1 dev enp0s9f0 table 200 proto static src 192.168.6.28 default via 192.168.2.35 dev enp0s18 table 201 proto static src 192.168.2.28 default table 250 proto static default via 192.168.2.35 dev enp0s18 proto static metric 100 nexthop via 192.168.2.35 dev enp0s18 weight 10 nexthop via 192.168.6.1 dev enp0s9f0 weight 75 [root@alex ~]#

We can see immediately :-
the default route for 192.168.6.28 is 192.168.6.1
the default route for 192.168.2.28 is 192.168.2.35
the weights for each WAN interface
the default route is 192.168.2.35 which is the only routing netstat shows...

Also you haven't indicated how your multi-wan is setup - primary, backup or standby?
Maybe a small sample from /var/log/syswatch when it fails would help us - from when you make the change to the first ping failure...
The reply is currently minimized Show
Accepted Answer
Nick Howitt

Offline
Sunday, January 29 2017, 02:52 PM - #Permalink
Resolved

0 votes

@Camhur ata,
Your problem is probably unrelated. Your driver, e1000, has been around for ages. It works fine for me in my mickey mouse VM (in VirtualBox using the same MT adaptor), but my VM is not set up in any workable way as both interfaces are natted.

There used to be an issue where syswatch would not work correctly if both WAN's had the same gateway IP which may be the case if they both have the same ISP.

If it is a VM related problem, I don't know how to troubleshoot.
The reply is currently minimized Show
Accepted Answer

Cumhur ata

Offline
Sunday, January 29 2017, 09:40 AM - #Permalink
Resolved

0 votes

Hi;

I really do not know where to look I read so many page and document but I still do not know what the problem might be
Maybe I have the same driver issue. Please make me sure I do not have driver issue. How can i check it? ClearOS works in vmWare(ESXI 5.5) MyClearOS version is
ens35 = LAN
ens33 = WAN2 (Same ISP with Static IP)
ens36 = WAN2 (Same ISP with Static IP)

Default route is ens33 when i change default gw to ens36 there is no internet but LAN works great.

[root@gateway ~]# uname -r
3.10.0-327.36.3.v7.x86_64
[root@gateway ~]#

[root@gateway ~]# lspci -k | grep Eth -A 3
02:00.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 01)
Subsystem: VMware PRO/1000 MT Single Port Adapter
Kernel driver in use: e1000
02:01.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 01)
Subsystem: VMware PRO/1000 MT Single Port Adapter
Kernel driver in use: e1000
02:03.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 01)
Subsystem: VMware PRO/1000 MT Single Port Adapter
Kernel driver in use: e1000
02:04.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 01)
Subsystem: VMware PRO/1000 MT Single Port Adapter
Kernel driver in use: e1000
02:05.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 01)
Subsystem: VMware PRO/1000 MT Single Port Adapter
Kernel driver in use: e1000

Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
default 192.168.1.1 0.0.0.0 UG 0 0 0 ens33
10.8.0.0 10.8.0.2 255.255.255.0 UG 0 0 0 tun0
10.8.0.2 0.0.0.0 255.255.255.255 UH 0 0 0 tun0
10.8.10.0 10.8.10.2 255.255.255.0 UG 0 0 0 tun1
10.8.10.2 0.0.0.0 255.255.255.255 UH 0 0 0 tun1
192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 ens33
192.168.2.0 0.0.0.0 255.255.255.0 U 0 0 0 ens35
192.168.3.0 0.0.0.0 255.255.255.0 U 0 0 0 ens36
The reply is currently minimized Show

Accepted Answer

Tony Ellis

Offline

Saturday, January 28 2017, 03:48 PM - #Permalink

Resolved

0 votes

This is one of my machines - one of the LAN connections is in fact a VLAN



[root@danda ~]# cat /etc/clearos/multiwan.conf

MULTIPATH="on"

MULTIPATH_WEIGHTS="eth3|50 eth1|10"

EXTIF_BACKUP=""

[root@danda ~]# cat /etc/clearos/network.conf

# Network mode

MODE="gateway"



# Network interface roles

EXTIF="eth1 eth3"

LANIF="eth0 eth2 eth0.1"

DMZIF=""

HOTIF=""



# Domain and Internet Hostname

DEFAULT_DOMAIN="sraellis.com"

INTERNET_HOSTNAME="danda.poweredbyclear.com"



# Extra LANS

EXTRALANS=""

[root@danda ~]# ifconfig

eth0      Link encap:Ethernet  HWaddr 68:05:CA:10:BF:48  

          inet addr:192.168.1.26  Bcast:192.168.1.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

          RX packets:72308703 errors:0 dropped:0 overruns:0 frame:0

          TX packets:122161302 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:1000 

          RX bytes:67292069059 (62.6 GiB)  TX bytes:159718505235 (148.7 GiB)

          Interrupt:17 Memory:f7bc0000-f7be0000 



eth0.1    Link encap:Ethernet  HWaddr 68:05:CA:10:BF:48  

          inet addr:192.168.5.26  Bcast:192.168.5.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

          RX packets:9572910 errors:0 dropped:0 overruns:0 frame:0

          TX packets:9567146 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:0 

          RX bytes:2098315386 (1.9 GiB)  TX bytes:70046876025 (65.2 GiB)



eth1      Link encap:Ethernet  HWaddr 68:05:CA:10:AF:A4  

          inet addr:192.168.2.26  Bcast:192.168.2.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

          RX packets:19297763 errors:0 dropped:0 overruns:0 frame:0

          TX packets:16097288 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:1000 

          RX bytes:10254731418 (9.5 GiB)  TX bytes:11136596745 (10.3 GiB)

          Interrupt:16 Memory:f7ac0000-f7ae0000 



eth2      Link encap:Ethernet  HWaddr 68:05:CA:0D:9D:C7  

          inet addr:192.168.3.26  Bcast:192.168.3.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

          RX packets:35481940 errors:0 dropped:0 overruns:0 frame:0

          TX packets:32471315 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:1000 

          RX bytes:6336301608 (5.9 GiB)  TX bytes:13629825485 (12.6 GiB)

          Interrupt:17 Memory:f78c0000-f78e0000 



eth3      Link encap:Ethernet  HWaddr 90:2B:34:3E:34:99  

          inet addr:192.168.6.26  Bcast:192.168.6.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

          RX packets:8787411 errors:0 dropped:0 overruns:0 frame:0

          TX packets:8787888 errors:0 dropped:0 overruns:0 carrier:5

          collisions:0 txqueuelen:1000 

          RX bytes:5700876954 (5.3 GiB)  TX bytes:6425368947 (5.9 GiB)



lo        Link encap:Local Loopback  

          inet addr:127.0.0.1  Mask:255.0.0.0

          UP LOOPBACK RUNNING  MTU:65536  Metric:1

          RX packets:3772487 errors:0 dropped:0 overruns:0 frame:0

          TX packets:3772487 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:0 

          RX bytes:4167926661 (3.8 GiB)  TX bytes:4167926661 (3.8 GiB)

The reply is currently minimized Show

Accepted Answer

Cumhur ata

Offline
Saturday, January 28 2017, 03:22 PM - #Permalink
Resolved

0 votes

I have the same problem.

is there any who can give us a sample that How MultiWAN Configuration of CLearOS has to be with 2 WAN and 3 LAN
The reply is currently minimized Show
Accepted Answer
Nick Howitt

Offline
Sunday, March 06 2016, 02:47 PM - #Permalink
Resolved

0 votes

There is nothing wrong with the RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller as long as you install the correct driver and make sure it is being used afterwards. Remember you may also need to either install a kmod-r8169 driver or blacklist the r8169 driver to make sure the r8168 driver is used.
The reply is currently minimized Show
Accepted Answer
Tony Ellis

Offline
Tuesday, March 01 2016, 12:34 AM - #Permalink
Resolved

0 votes

You definitely will get flaky behaviour with the incorrect driver. Difficult to know whether it will completely fix your problem, but would sure help... You can change the two ping addresses by editing the file "/etc/syswatch" - see the bottom few lines. An example :-

..... ..... # Ping servers # - The syswatch daemon by default will just ping the next hop on the Internet # and another device. You can override this with a comma separated list of # 2 servers. #ping_servers=111.222.333.444,555.666.777.888 ping_servers=8.8.8.8,8.8.4.4

Like

1

Steve Thompson likes this post.
The reply is currently minimized Show
Accepted Answer

Steve Thompson

Offline
Monday, February 29 2016, 10:25 PM - #Permalink
Resolved

0 votes

I am having a similar problem with ClearOS 6.5 (kernel 2.6.32-573.1.1.v6). I talked with my one ISP (Metronet, Fiber optic...) and their person told me that their "routers" do not have a hi-priority in responding to PINGs. And I'm given to understand after this past weekend's maint by the local RR vendor (BrightHouse Networks) that I can get into this situation where eth3 is the only working WAN (RR) and then it ping-ponged to eth0(Metronet).

What was suggested is using a different heartbeat address. Well, I'm not given that as an option, and in reading the online doc, I don't know that it is a config option.

Meanwhile, it appears that I have the dreaded LAN controllers: RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller.

Would you think that just putting the updated drivers in/on would solve this, or should I go even further to fix the system to use something else to ping?

What clued me in to the real problem I had was that I could not get things to reliably resolve. And a ping of 8.8.8.8 was ignored. And this was happening while trying to fix a vbox problem on a SUSE 13.2 system.

So, suggestions?
The reply is currently minimized Show
Accepted Answer
Nick Howitt

Offline
Thursday, February 25 2016, 08:10 AM - #Permalink
Resolved

0 votes

Please see this post.
The reply is currently minimized Show
Accepted Answer
Nick Howitt

Offline
Monday, February 22 2016, 05:50 PM - #Permalink
Resolved

0 votes

If you disable syswatch, you disable your failover. To answer you original question now, you can change the IP's used by syswatch in /etc/syswatch but you need to use something appropriate. I thought ClearOS used your internet gateway first, which you cannot (should not?) specify as an override as each interface has a different gateway IP. I'm not sure how it works in a MultiWAN environment.

Can you give the output to:
lspci -k | grep Eth -A 3
Just to check the right drivers are being used now.
The reply is currently minimized Show
Accepted Answer

Michael Rodriguez

Offline
Monday, February 22 2016, 05:21 PM - #Permalink
Resolved

0 votes

I tried the new drivers and it seems to have lessened the effect but it is still dropping the connection due to ping loss. Is there a way I can disable syswatch or a work around?

Mon Feb 22 11:05:05 2016 info: system - heartbeat...
Mon Feb 22 11:08:49 2016 info: enp1s0 - ping check on server #1 failed - 8.8.8.8
Mon Feb 22 11:08:54 2016 info: enp1s0 - ping check on server #2 failed - 54.152.208.245
Mon Feb 22 11:08:54 2016 warn: enp1s0 - connection warning
Mon Feb 22 11:09:06 2016 info: enp1s0 - ping check on gateway failed - 75.148.200.246
Mon Feb 22 11:09:08 2016 info: enp1s0 - ping check on server #1 failed - 8.8.8.8
Mon Feb 22 11:09:13 2016 info: enp1s0 - ping check on server #2 failed - 54.152.208.245
Mon Feb 22 11:09:13 2016 warn: enp1s0 - connection warning
Mon Feb 22 11:09:25 2016 info: enp1s0 - ping check on server #1 failed - 8.8.8.8
Mon Feb 22 11:09:30 2016 info: enp1s0 - ping check on server #2 failed - 54.152.208.245
Mon Feb 22 11:09:30 2016 warn: enp1s0 - upstream connection problems with your ISP
Mon Feb 22 11:09:30 2016 info: system - changing active WAN list - enp2s0 (was enp1s0 enp2s0)
Mon Feb 22 11:09:30 2016 info: system - current WANs in use - enp2s0
Mon Feb 22 11:09:30 2016 warn: system - default route is not active... removing - enp1s0
Mon Feb 22 11:09:30 2016 info: system - restarting firewall
Mon Feb 22 11:09:34 2016 info: system - default route vanished - using enp2s0
Mon Feb 22 11:09:34 2016 info: system - setting default route to 10.1.10.1
Mon Feb 22 11:09:34 2016 info: system - restarting firewall
Mon Feb 22 11:09:37 2016 info: system - restarting firewall
Mon Feb 22 11:09:53 2016 info: enp1s0 - ping check on server #1 failed - 8.8.8.8
Mon Feb 22 11:09:56 2016 info: enp1s0 - ping check on server #2 passed - 54.152.208.245
Mon Feb 22 11:09:56 2016 info: system - changing active WAN list - enp1s0 enp2s0 (was enp2s0)
Mon Feb 22 11:09:56 2016 info: system - current WANs in use - enp1s0
Mon Feb 22 11:09:56 2016 warn: system - default route is not active... removing - enp2s0
Mon Feb 22 11:09:56 2016 info: system - restarting firewall
Mon Feb 22 11:10:00 2016 info: system - default route vanished - using enp1s0
Mon Feb 22 11:10:00 2016 info: system - setting default route to 75.148.200.246
Mon Feb 22 11:10:00 2016 info: system - restarting firewall
Mon Feb 22 11:10:04 2016 info: system - restarting firewall
Mon Feb 22 11:10:28 2016 info: system - heartbeat...
Mon Feb 22 11:10:28 2016 info: enp1s0 - ping check on server #1 passed - 8.8.8.8
Mon Feb 22 11:15:29 2016 info: system - heartbeat...
The reply is currently minimized Show
Accepted Answer
Nick Howitt

Offline
Tuesday, February 16 2016, 07:21 PM - #Permalink
Resolved

0 votes

Hi Michael,
Your reply will get through when the mods approve it. I think it may be the last one requiring approval.

For the RTL NIC, please install this driver. For the 82574L please install this one. To install do a "yum localinstall --nogpgcheck ..." or "rpm -ivh ...". Reboot once they are installed.

In case you are wondering what the drivers are, they are the kmod drivers supplied by ElRepo which I've compiled for ClearOS. They are based on the original manufacturers' drivers but changed so they should not need recompiling every time the kernel goes through a minor update.
The reply is currently minimized Show
Accepted Answer

Michael Rodriguez

Offline
Tuesday, February 16 2016, 05:50 PM - #Permalink
Resolved

0 votes

alright let me try this.
The reply is currently minimized Show
Accepted Answer
Nick Howitt

Offline
Tuesday, February 16 2016, 05:35 PM - #Permalink
Resolved

0 votes

Let's fix some underlying problems first. Is Comcast your ISP for both WAN connections? If so, do you know if their DNS servers are available for general public use (i.e. to people not connected to their network)? They appear to be which is good in a multiwan set up.

You at least have an issue with the RTL8111/8168/8411 driver, but to fix it I need the output of "uname -r" which should tell me the version of ClearOS you're running.

You may also have an issue with the 82574L driver. For this, what is the driver version reported by "modinfo e1000e"?

By bad luck you have two of the most common adaptors with driver issues, but fixing them should be easy.
The reply is currently minimized Show

Your Reply

Please login to post a reply

You will need to be logged in to be able to post a reply. Login using the form on the right or register an account if you are new here.

Community Forums

ClearOS Portal

ClearVM Platform

ClearVM 2 Platform

Forums