Jump to content
Welcome to our new Citrix community!
  • 0

citrix workspace app 18.10 floods dns servers with "No server" lookups 40gb dns traffic per week


Michiel de Groodt

Question

Hello,

 

We are having problems with the Citrix workspace app 18.10 for linux. When starting a connection to our storefront we get 20 "no server" dns queries being send to our dns servers every second per device. This does not happen in citrix receiver 10.* nor in 18.08. This happens on IGEL laptops with 18.10 and hp thinpro t510 thin clients. I have not been able to reproduce this on the windows worksapce app 18.10.

We have 430 devices and this generates 43GB dns traffic a week. If i turn on logging of the firewall rule for dns the log fils up to 9.6 GB within a day.....


3/26/2019 11:23:20 AM 0F54 PACKET  0000004B10904180 UDP Rcv 10.100.12.12    c6e8   Q [0001   D   NOERROR] AAAA   (11)no%20server(0)

3/26/2019 11:23:20 AM 0F54 PACKET  0000004B10904180 UDP Snd 10.100.12.12    c6e8 R Q [8281   DR SERVFAIL] AAAA   (11)no%20server(0)

3/26/2019 11:23:20 AM 0F54 PACKET  0000004B10EA4120 UDP Rcv 10.100.12.12    c6e8   Q [0001   D   NOERROR] AAAA   (11)no%20server(0)

3/26/2019 11:23:20 AM 0F54 PACKET  0000004B10EA4120 UDP Snd 10.100.12.12    c6e8 R Q [8281   DR SERVFAIL] AAAA   (11)no%20server(0)

3/26/2019 11:23:20 AM 0F54 PACKET  0000004B12E10B90 UDP Rcv 10.106.10.3     80f2   Q [0001   D   NOERROR] A      (11)no%20server(0)

3/26/2019 11:23:20 AM 0F54 PACKET  0000004B12E10B90 UDP Snd 10.106.10.3     80f2 R Q [8281   DR SERVFAIL] A      (11)no%20server(0)

3/26/2019 11:23:20 AM 0F54 PACKET  0000004B10904180 UDP Rcv 10.106.10.3     aca5   Q [0001   D   NOERROR] A      (11)no%20server(0)

3/26/2019 11:23:20 AM 0F54 PACKET  0000004B10904180 UDP Snd 10.106.10.3     aca5 R Q [8281   DR SERVFAIL] A      (11)no%20server(0)

this goes on and on and on

 

I cant find this to be a known issue. Anyone else experiencing this?


We run 7.15CU3

WIndows 2012R2DTC

Storefront 3.12.3

We do not use a netscaler for internal connections.

 

Gr,


Michiel

 

Link to comment

5 answers to this question

Recommended Posts

3 minutes ago, BART JACOBS1709152229 said:

Please have a look at this:

 

https://support.citrix.com/article/CTX206182

 

Part of this behavior is "by design". But the amount is certainly not. 

Did you try version 1901?

I allready looked at it thanks. But this is not the same. I see the ones from above as well and those are fine but not the "No Server" ones. Also notice mine are SERVFAIL

Cant upgrade since neither IGEL nor HP has released a higher workspace version then 18.10

Link to comment

I can confirm this is not just an issue with the thinclients. I installed 18.10 as well as 19.01 on a ubuntu virtual deskop (both times in a clean state) and the problem remains.

 

We are getting a total of 650 000 000 dns queries a week from 430 devices........ Currently in contact with a citrix engineer investigating this.

Anyone else experiencing this?

Turn on debug logging on your dns servers for a minute and check the log for "SERVAIL" RCODE2

Link to comment

I have a workaround for this. Please follow the below steps(No need to do the workaround for every session launch via web. Doing it once for one store is enough).

 

1.       Login to SSP(Native receiver) and add the store via UI.

(  If the client is not in the same domain, you have to trust the CA certificate - https://support.citrix.com/article/CTX231524?recommended

And

/etc/resolv.config – file has to be updated with the DNS IP

Ex:

nameserver 10.120.16.17

search xyz.net )

2.       Logoff form native receiver.

3.       Kill the below processes

a.       Killall AuthManagerDaemon

b.       Killall ServiceRecord

4.       Launch the store via web and launch the session.

 

Now the No server OR ServerFail message will not be there. Because, Step-1 is adding the store URL to “ServiceRecord.xml” and it will be able to ping the server. Whenever there is a “ServerFail” message the application is trying to ping the server recursively so its flooding the network.

 

In simple words :: We have to have a valid store URl in the "ServiceRecord.xml" file.

(xml file location : /home/currentuser/.ICAClient/config/ServiceRecord.xml)

The “.ICAClient” folder is hidden.

Link to comment

Here at Stratodesk we have had similar reports from our customers too for our NoTouch OS with Citrix Workspace App 18.10. An easy workaround that we found for those who were affected is to make a predefined host entry, mapping no%20server to 127.0.0.1 (see screenshots). In principle this should work on other endpoints too. The latest receiver we have in GA/production NoTouch OS  is 19.3 (not yet assessed if the workaround is still necessary)

 

 

Screen Shot 2019-04-18 at 11.57.34 AM.png

Screen Shot 2019-04-18 at 11.57.24 AM.png

Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...