Help Center

Follow

Computer offline

**Edited 1.25.2016**

Description

The target computer is offline and the system's Offline Settings prevents deployment to offline computers.

Possible Causes

A computer's offline status is determined by ping (ICMP echo).  If the computer is online but not responding to pings it will be treated as though it is offline.

Recommended Fix

Correct the issue preventing the pings from responding (such as a firewall) or change the Offline Settings (File > Preferences > Deployments) to allow use Ping for Offline Testing or Send Wake-on-LAN to Offline Computers.

 

Available Offline Settings:

2016-01-25_13-19-04.png

Was this article helpful?
0 out of 1 found this helpful
Have more questions? Submit a request

13 Comments

  • 0

    I know I've asked about this before but it is still pestering me. You say pdq determines offline status by ping. Well both the server (pdq installed) & the client (pc I want software deployed to) can ping each other with window's ping command in cmd line. Yet pdq continues to say it is offline...

  • 0

    The ping is done using the fully qualified domain name of the computer, which may not be resolving correctly.  This is usually the cause of the difference, but if you can ping using the full name then there would be another issue that needs to be looked into.  It could also be a timeout, PDQ Inventory uses a 1 second timeout (approximately) and the computer may be taking too long to respond.

  • 0

    Whenever I ping, the FDQN is automatically appended. I can also ping by FDQN to begin with. The ip address for either ping is correct & ping is successful. It might be the timeout within pdq that you referenced because after about 3-5 minutes pdq finally decided to believe the computer was online.

  • 0

    Interesting, thank you.  It may also be the DNS cache, particularly if it's a computer that has recently come online.  The cache does seem to be somewhat process dependent so that the once a process has failed to resolve a computer it will continue to do so for a while even after the computer becomes available in DNS. 

    We're working on some other issues similar to this, where the service process has issues pinging after a while.  Hopefully those workarounds will help out here as well.

  • 0

    Could it be, that you forgot to set up you firewall so it allows PDQ Deploy to connect?

    Otherwise try CMD > ipconfig /flushdns and ipconfig /registerdns

  • 0

    No, client firewalls are off.

    I have tried flush on both server & client. Don't recall trying register. Will try next time.

  • 0

    Jonathan,

    When you notice the computer is offline in PDQ Inventory but you can ping it, do you run the heartbeat manually or wait for the automatic heartbeat?

  • 0

    I run it manually & it still stays offline. Eventually (depening on how much time passed) it will either automatically find it online or I just happened to manually do it at the right time that it will find it online.

    @ SelfMan - thinking it twice, I don't think register would do much since flushing has removed the entry from cache & when pinged it returns the correct ip but I will try it when I come across it again.

  • 0

    Definitely sounds like a DNS cache issue. 

  • 0

    Yes, I believe the last time I asked this question this was the conclusion that was reached. Just not sure how it's a cache issue when the cache has been flushed & the right ip is returned on a new lookup / ping.

  • 0

    The cache seems to maintain some information per process, so the service proces which performs the heartbeat still has some dirty cache data even after you flush it.  The issue actually seems to be with the Ping API, it seems to maintain some of its own cache data separate from the main DNS cache.

    One thing you can try is that when you see it again restart the PDQInventory service.  If it starts pinging correctly after a restart it would seem to be something cached within the cache process.

  • 0

    This ping problem has been seen in PingInfoView from Nir Sofer, who himself decribes it as a mystery.

    http://www.nirsoft.net/utils/multiple_ping_tool.html

    See known issues section and change log. What I can tell, he managed a way to fix it or create a workaround.

  • 0

    On your DC, check your forward look up zone for your domain. I was having a similar issue and a local flush and register wasn't cutting it so I went in there and noticed that computer entry was outdated and registered to a different IP address than it actually was. So I deleted its entry in DNS, rebooted the target PC, it automatically registered with the DNS service, reran a Standard scan and all was good. Hope this helps.

Article is closed for comments.