Lost Sync wit 5.0.0.7

[LordCrush]

Hi all,

my very loved "lost Sync" Error is back. SEQ 5 does loose sync after a short time in a Zone .. initial decode and player position is ok, but all new spawns come up as unknown


Debug: SEQ: Giving up on finding arq 2be6 in stream zone-client cache, skipping!
Debug: SEQ: Giving up on finding arq 2cb2 in stream zone-client cache, skipping!
Warning: Lost sync, relog or zone to reset
Warning: uncompress failed on 0x71da: -3 - data error
no further attempts will be made until zone on stream zone-client.

Btw is there an option to switch off the info-output on the console and reduce it to debug and warning ?

Thank you in Advance

-- Lord Crush

[Pascal7]

Had the same problem with 5.0.0.7 so I downgraded to 5.0.0.6 same problem.



Debug: SEQ: Giving up on finding arq 4cd6 in stream zone-client cache, skipping!
Warning: EQPacketFragmentSequence::addFragment: WARNING OpCode 0x5a50 will not be processed due to loss
Warning: EQPacketFragmentSequence::addFragment(): recieved Out-Of-Order fragment seq 0x0006 (0x0009) expected 0x0005


I am using the new spells file. Going to try to wipe out my SEQ directory and start fresh. Maybe there was a rement of 5.0.0.7 in there.

[Pascal7]

Pulled out my old SEQ box that I used up to 5.0.0.6 and had the spells_us.txt prior to the last patch. Oddly enough it doesn't error our like my SEQ Laptop with the new spells file.

Will do some more testing tonight with both hooked up.:confused:

EDIT::
Tried swapping back to the old files. Nada.. :(
So I reinstalled gdbm, autoconf, automake, and did a make distclean and reinstalled 5.0.0.7... /groan




Zone: EntryCode: Client
Warning: Lost sync, relog or zone to reset
Warning: uncompress failed on 0x523e: -3 - data error
no further attempts will be made until zone on stream zone-client.


Now I'm not decoding anything, before I was at least getting the zone identified. Guess I'll try again on the old laptop here. Really wish I had taken some more networking classes now.. /bonk self.

[Omiime]

Originally posted by Pascal7
Had the same problem with 5.0.0.7 so I downgraded to 5.0.0.6 same problem.



I am using the new spells file. Going to try to wipe out my SEQ directory and start fresh. Maybe there was a rement of 5.0.0.7 in there.


Had this problem for very long time. has nothing to do with 5.x series. I've forgotten what caused it tho. Packets dropping while sniffing or something ?

[Pascal7]

Had nothing to do with SEQ. :o

My NIC decieded to take a partial dump, replaced PCMCIA NIC with a shiney new D-Link one and Bammo SEQ working like a champ again. :D

I think that I may have killed the NIC when I upgraded.

[LordCrush]

It seems that i have the problems due to too much collisons on the Hub, beause i run 2 or 3 SEQ-Sessions and use VNC to remote control the Sessions

/shrug

[Cryonic]

Spend a few bucks and install a second nic in each SEQ box. Hook the new nic up to the switches and use that nic to do VNC and the one to the hub to do sniffing.

[S_B_R]

I agree with Cryonic. In the other thread you tried changing the topology to no evail. So putting the VNC traffic onto the switch would be your best bet, unfortunately it's not a free fix. ;) There's always trying to Optimize VNC to consume less bandwidth, but to keep VNC as a useful way of displaying SEQ you're kind of limited on options.

[LordCrush]

I have no problem with a second NIC ... there should be more than enough in my box with cards... But how do i tell VNC to use the right NIC ... with SEQ i can configure the interface

That would look like this




Internet
|
Router
|
|
--------------HUB------------------------------
| | eth0
| SEQ-Box
| | eth1
-----------Switch---------------------------------
|
EQ-Box ....



Do i need an IP on eth0 to capture packets ? or can i use any, but that for my Segment ?

i.e 10.50.200.0/24




Internet
|
Router
|
|
--------------HUB------------------------------
| | eth0 (10.50.201.1) dummy and SEQ Watch NIC
| SEQ-Box
| | eth1 (10.50.200.2) "Work" NIC for VNC
-----------Switch---------------------------------
|
EQ-Box (10.50.200.22)



Thanx in advance :)

[Dedpoet]

In your first picture, you would just need to make sure eth1 is your default gateway, so by default, all packets (vnc included) will use that gateway. To be honest, it's probably easier if that NIC is eth0 (the default, default gateway if you will). Then you just need to make sure your NIC is in promiscuous mode and run showeq to capture the packets:

# ifconfig eth1 promisc up
# showeq -i eth1

Assuming that eth1 is the one that's attached to your hub.

[S_B_R]

According to your topology, eth0 is connected to the hub, and eth1 is connected to the switch. Here's what you do. Configure eth1 with an ip and gateway information for your network or DHCP if needed. Then
ifdown eth0
ifconfig eth0 promisc up then just run SEQ as normal with "-i eth0" as an option.

[LordCrush]

Thank you all for you suggestions :D!!

With the second nic - only for SEQ - it works now fine :)

[S_B_R]

Glad to be of service ;)