Forum Discussion
CirrusAble to telnet to virtual when the pool members are down ?
Hello Folks,
I am trying to understand a strange behaviour seen on the LTMs.
Configs
---------------------------------------------------
virtual abdc {
pool abdc
destination 1.1.1.1:http
ip protocol tcp
rules SNAT2VIP
profiles {
http-xff {}
tcp-lan-optimized {
serverside
}
tcp-wan-optimized {
clientside
}
}
}
+-> POOL MEMBER abdc_80/2.2.2.2:http inactive,down
+-> POOL MEMBER abdc_80/3.3.3.3:http inactive,down
-------------------------------------------------------------------------------------------
Question :
1.) why is that i am able to successfully telnet to the virtual ip ?
2.) Is the bigip not suposed to send a reset packet when all the pool members are down ?
3.) I have a GTM polling this LTM virtual with a tcp_half_open monitor along with bigd , the GTM sees the virtual to be UP though the virtual is down on the LTM. ( more of tied to the 1st question )
Please help me in understanding this
Ps
( i know its not right to mix the bigd and tcp monitors for a LTM virtual on the GTM..)
3 Replies
- nitass
Employee
have you seen this one?
sol8082: Overview of TCP connection set-up for BIG-IP LTM virtual server types
http://support.f5.com/kb/en-us/solutions/public/8000/000/sol8082.html
hope this helps. 
What_Lies_Bene1Cirrostratus
Roger, just in case you don't find that document clear enough I'll quickly break it down for you;
1) You can telnet to the Virtual Server even with the Pool Members down. This is the full proxy in effect, the BIG-IP is terminating your connection and will establish a different one to a Pool Member when it needs to.
2) The TCP three way handshake is completed but the moment you send a HTTP request on this connection, you'll then get the RST. This is normal behavior for a Standard Virtual Server with a HTTP Profile assigned.
3) Use a HTTP monitor (that sends a HTTP request) and this won't occur.
bcrogerzPerfect! it makes perfect sense now !
Many thanks both of you
Roger
