NTP (Network Time) Appliance

[VMTN_Admin]

http://www.vmware.com/vmtn/appliances/directory/210

A hardened appliance, with Windows support, that will communicate with NTP servers around the world to keep current time.

[Dennis_Parker]

Appliance built with an IDE drive so it can't be used in an ESX server environment. Makes this much less usable.

[jss27560]

I installed the VM and the time is not correct. It shows the Universal time as what my local time is and the local time as 4 hours earlier when it should be 4 hours later.

[kamphor]

Thank you for using my virtual appliance. Unfortunately, this appliance was published for the Summer 2006 VMware Challenge. The only virtualization software available \*for* the Challenge was Workstation and GSX/Server. Thus, this appliance was specifically tailored for them and not for ESX server. I thoroughly tested obtaining correct time with all time zones before I published my entry into the contest. If an option exists in the vmware software for time synchronization between the host and guest machine, you may want to disable that function.

I lost all my development work for my challenge entry due to a hard disk failure. However, all that really needs to be done is to recompile a linux kernel specifically for the ESX virtual hardware and replace the one that boots in the ide configured virtual machine. I do not provide support for this appliance and wish you luck in any modifications you make to the appliance. For example, you are free to use whatever scripts I wrote and integrate them into an existing linux/unix box.

-Kamphor

[eeAcn]

Hello,

Is there a recommendation for security or kernel patches for this appliance? Is patching possible or recommended for this appliance?

Regards,

Eric

[mfwade]

Can someone please provide me a link for the download. The TORRENT doesnt seem to work and I am really interested in this VM.

Thank you.

[Goatie]

Hi,

I was interested in this until I too realised it was an IDE disk.

All VMware products allow the use of SCSI disks (Workstation, GSX and Server) and making your appliance on a SCSI disk allows it to be used in ESX as well.

Could you please rebuild it onto a SCSI disk so it is usable by the widest of users?

All appliances should be done on SCSI and not IDE.

Cheers,

Steve.

[RDPetruska]

>All appliances should be done on SCSI and not IDE.

Actually, this is one of my main arguments for adding IDE disk support to ESX. I can't believe there is any technical reason they haven't done so. And now, with the big push for Virtual Appliances, as well as VDI - I don't understand why they haven't added IDE support yet.

[Goatie]

Totally agree. It has screwed us up a bit. When we first started on the virtualisation track we had a server running VMware Workstation to try things out a bit. One of the guy's created a bunch of VMs, but used IDE disks so we had to rebuild them when we went to ESX.

Although IDE's already in there!!! It's the CDROM!

Most of the code's in there, they have just not made it available the extra step of making it a hard disk.

[RDPetruska]

Add your voice to this thread in the feature request forum, then... http://www.vmware.com/community/thread.jspa?threadID=62873&tstart=0 , and include a link to this thread for more concrete evidence of good reasons TO include the support in their Big Buck$ program.

[penzipa]

Has this been ported over to scsi or does anyone know of another virtual machine that can be used as an ntp server

[jhicks]

I'm currently debating placing our NTP source services on a VM. The system currently hosting NTP (and many other services) is a physical box, which is being replaced by a VM. I'm reluctant to run our time source on a VM due to time drift in the virtual world. How is time drift being addressed by this virtual appliance?

Or has time drift been adequately addressed within ESX, making this no longer a concern?

[TylerApache]

Apparently another "GOTCHA" with this appliance is that it has the old Daylight Savings rules in it. Anyone know how to update this appliance?

[kamphor]

I posted an updated and stable version that should hopefully address the issues everyone has been having. It's posted in the community virtual appliances section.

or go directly to the page here- http://www.trawick.com/ntpvmapp.html

Thanks for the comments and feedback!

-Mervin Yap

[kamphor]

Hi all!

Just wanted to mention that there is a small problem with the appliance. After a few days of running, the ntpd process may kick out and stop. I will try to make a fix for this in the next few days and post an update for all 3 versions of the appliance.

I'll keep everyone posted on this discussion thread when the patch is applied and ready to download the tar balls.

-Mervin Yap

[jhicks]

Excellent! Good to see an updated appliance that will run on ESX. You briefly mention in your notes "more frequent polling" to address time drift issues. Is this the only method you're employing to keep time stable on the VM? Or are you also using any timeTracker options in the vmx config and/or other active processes within the VM?

Thanks & good luck tracking down and resolving the current issue you've found...

J

[kamphor]

A work around was applied to all versions. Please redownload!

I also supplied an ESX3 OVF archive.

let me know if other issues arise.

-Mervin Yap

[magnetron]

Kamphor, kudos for getting across a seemingly great product, I have two queries:

Will the NTP work on free VM Player and Current free VM server 2.0 offering, (not the ESX)?

Do I need to install OVL file over the original or will only the OVL file be sifficient?

Pls advise!!

[kamphor]

magnetron,

if you go to the download page, I believe the ide w/vmxnet version should work fine with the free vmplayer. The scsi version w/e1000 should work fine with the free vmware server 1.0.x. I have yet to test it with vmware server 2.0. If anyone has a test environment for vmware server 2.0, it would help greatly if someone can supply feedback on testing it in that version.

respectfully,

Mervin Yap

[magnetron]

Thanks for the reply Kamphor. Do you suggest that if the NTP is hosted on VM

player it would be most resource appropriate and effective on a desktop

class base machine?

Also, if I host NTP server within my DMZ to serve my firewalls and routers

too. I would only need to use single port through my firewall towards

internal LAN on hosting windows 2003 servers, is this the right design?

Pls advise!

[kamphor]

hicksj,

The purpose of the ntpd daemon is to poll the time servers and adjust the clock accordingly. In a real machine, this works fine. However, under vmware, the clock seems to drift a lot. In this case, I force a time adjustment via the ntpdate program instead of the usual ntpd time slewing method. I have the cron daemon sync up every minute. In addition, I wrote a small script that will check to see if ntpd is running. If it is not, it will restart itself (which also kick starts another force time adjustment). This check is also done every minute. One more script was made that sync'd the system clock to the hardware clock (guest vm clock in nvram file?) every 2 seconds. Already part of the original design, the vmx file is configured not to sync the guest virtual machine with the host system clock. All these work arounds should hopefully make the system more stable.

Before the workarounds, the ntpd process may kill itself because the time offset/jitter surpassed a certain millisecond threshold. The windows "net time" command would still work. However, the clock will drift significantly and no clients can communicate with the ntpserver via udp/tcp port 123 since the process is killed.

Other than that, I cleaned up the generation of the ntp.conf file as I noticed duplicate ips from time to time.

So, I hope all these fixes essentially makes the vm app a true "final" release =).

Thanks all for the comments and bug notices!

-Mervin Yap

[loetmann]

Hi,

first of all, thanks for this small but great virtual app.

Since setting up a reliable NTP-Server is a hard work, I was really exited stumbling upon this VM.

It was my first time install on VMServer 1.0.3:

• I tried the IDE version first, but several processes complained about read-only access to filesystem

• So I switched to the SCSI-Version: Everything went great!

• After syncing some windows machines with their new ntp - Master I tried so with my debian and ubuntu boxes:

bq. ntpdate 172.16.117.226
5 Feb 11:19:35 ntpdate[6685]: no server suitable for synchronization found

• So I had a look into shorewall rules and so on

• But linux boxes obviously not able to speak ntp with this VM

Any ideas ?

[kamphor]

loetmann,

Thanks for the update. I will investigate the ide version today. Since the new versions of VMware supports scsi, I'm wondering if I should just remove the ide version from the page?

Have your tried using the "ntpdate -u" flag? On a few of my linux boxes, the time won't sync unless I use the "-u flag".

Let me know if this works. If not, then I'll have to modify the ntp.conf generation script.

Thanks,

-Mervin Yap

[kamphor]

loetmann,

okay...i think I figured it out. It's like how the windows ntp client will only adjust time with a certain stratum number for the ntp server.

enable the status page and use alt-f12 to view the ntp peer stats. wait until you get an asterisk ( * ), plus ( + ), or minus ( - ) symbol before a server, then you should be able to use the ntpdate program on your linux/unix boxes to sync up the time. preferably wait for an asterisk or plus symbol. the ntpd daemon needs some time to "stabilize" (about 2 minutes) as it figures out what stratum it should configure itself to be.

Hope this helps.

-mervin yap

[loetmann]

kamphor,

first of all, thanks for your prompt reply and reaction. I had to figure out a lot of issues to get this VM up and running.

Obviously, as allready said before in this thread, RTC under VMWare is a major challenge.

I use to run VMServer 1.0.4 under 32-bit Debian etch with a 2.6.18 Kernel on a quad-XEON machine (Dell 860).

At the time I started evaluating your NTP-VM I ran the standard Server-bigmem kernel. Previously I solved my RTC and clock skew problems with some also debian-based guest systems with migrating to Ubuntu 7.10-JEOS, which is somewhat more performant.

Since I seriously wanted to give NTP-VM a chance I decided yesterday to take the bully by the horns in order to solve this RTC-issue permanently. With a little help of google and friends I figured out that there are two factors to be considered:

1. HPET_EMULATE_RTC has to be activated statical and not as a Module:

2. Changing timer frequency from 250Hz to 1000Hz ( as vmware even propagates, by the way)

This was necessary to avoid the heavy clock skew I encountered in NTP-VM and the boring messages on the server console: kernel: rtc: lost some interrupts at 512Hz.

Now NTP-VM didn't stumble any more behind. But after even waiting for hours ntp to become synchronized (stratum evaluation) I remembered another trick I used some time ago: kernel boot parameters in the guest machine.

After changing following parameters in menu.lst to:

- clock=pit

+ noapic nolapic nosmp

nearly everything is allright. NTP is evaluating stratum within minutes and the service becomes availlable also for *nix machines. Fine.

But there are three issues left:

I am not able to start up the VM without interacting with the console to give this starting udev thing a hit with CTRL-C. It allways stopped at this point. First I thought this was due to clock skew or clock frequency ratio. But no chance.

After your last update I tried again the scsi version. For me its now brocken. While uncompressing the boot image it dies. So I changed to IDE-version which now works fine with said modifications to server and guest.

Somewhere in your descriptions on vmnet is mentioned that vmware tools should be installed. My Server-console complains about missing vmware tools.

Since this VM doesn't have even a sshd running, I am not able to shut it down by remote scripting.

Excuse my epic description and thanks a lot for your work and support.

Regards

- Ralf Kuehnbaum-Grashorn

[kamphor]

Thanks for your input. Have you waited longer at boot time? You have to be patient as I have a 10second sleep time at "Starting udev." so you have to wait a little longer. If the system does not continue after 30 seconds or so, hit any key ("i" seems to work good).

I have already compiled the linux kernel in the guest system for 1000hz as well as setting preemption to low latency desktop mode. clock pit is also already set in the guest vm as a kernel boot option.

I'll investigate more on vmtools. it is installed in the guest vm already. if you go to console and do a: "chkconfig --list" it should be listed there, but let me know if it's not working properly.

I may have to redo this vm under a different distro. I've noticed boot lockups at/around "starting udev" in the past.

Thanks again...and I'll try to start converting/updating this system with a newer distro and kernel!

Keep me posted on any bugs you encounter...the more eyes on this appliance the better. and the better for the vm community!

-Mervin Yap

[loetmann]

Hello kamphor,

yes, I have waited definately more than the requested 10 sec.

Regarding clock pit there was a typo in my posting: I have removed "clock=pit" and added "noapic nolapic nosmp" in menu.lst

The vmware-console complains about missing vmwaretools. chkconfig --list says it should be on for init-levels 2 to 5.

I tried also shutting down the vm: It was a hard one or just pulling the power-cord.

I would apreciate in choosing a more actual distribution. If you are not sticked with mandriva, give ubunto Jeos a try. I ran it allready with a footprint of 32 MB for a secondary DNS.

-Ralf

[tasuser]

Ralf,

It sounds like you have the NTP appliance working well, or possibly have one working under Jeos. If so, would you consider submitting it as an appliance? There just aren't many dedicated NTP appliances to choose from. I would be willing, but I'm still VERY new to Linux.

Note to kamphor - I do appreciate all the work you have done thus far - I just can't seem to get it to work in my environment (ESX 3.02).

[Rob_Rowald]

I am having an issue with the appliance and hopefully you or someone else can help.

I have it imported into my ESX cluster and when i start it up it complains that it cannot find the e1000 adapter and that is is delaying initialization.

I tried changing the adapter type to vmxnet but that did no good, so i switched it back.

I went into the bash console and typed ifup eth0 and got the following response:

/bin/true

e1000 device eth0 does not seem to be present, delaying initialization.

Any help woould be appreciated

Thanks

Rob

[trhoads]

I just finished working out a few issues with the NTP Appliance on my 3.5.0 ESX server and wanted to share my fixes.

Be sure in your ntpvm_scsi.vmx (or whatever .vmx corresponds to your VM) that you have the line:

ethernet0.virtualdev = "e1000"

I know it's in the original ntpvm_scsi.vmx, but when you bring it into the host it can re-write the file.

-Tom

[trhoads]

Kamphor,

Thanks for doing this project, the appliance is killer. I looked around quite a bit and found that this udev bug has been around in Mandriva for some time.

In order to get around the hang at "Start udev", since we do not use wireless NIC's in our ESX servers:

edit /etc/modprobe.conf and add the following line:

"alias ipw2200 off"

-Tom

[jeaubain]

Hi I am trying to download the OVF version, however i am redirected to an FTP server that requires a password can someone please assist me with this issue.

Thanks in advance

[mvarre]

Hello, can anyone tell me if the UDEV startup problems have been definitively resolved? I have tried adding Tom's fix to /etc/modprobe.conf however i still continue to be required to interact with the VM in order to get it booted passed "Start udev".

I would love to get this going, but obviously I cant rely on a VM not fully starting on it's own if it needs to get rebooted.

Other than this one tiny (yet huge) issue, the VM is solid and I think its fantastic. Thanks!

[Halon330]

When booting the appliance never can resolve time servers to pull time from. Is there a way to configure what DNS servers the appliance uses. I tried adding two time servers manually by IP address, but I'm not sure that the appliance is using them. When I press altf12 to view status nothing is ever listed on the screen, and I cannot figure out how to get back to the menus. I have tried altf11 and alt+f12 to get out of the status screen.

Anyone have any advice here?

[slartimitvar]

I'd really like to have a look at this but here is currently a problem with ftp://kamphor.trawick.com

Did someone trip over the power cord?

[James_Bucknall]

Slartimitvar

Try this link. It has been working this morning..

Regards,

James

[JoeAK]

Thanks for a great applicance. I have a quick question, any way to add additional interfaces to the applicance? I have 3 different networks on my ESXi server, two of them just internal but I would like to use one applicance to provide clock to all 3 vSwitches. How would I go about adding additional network cards to the applicance?

Thanks

Haris

[einstein-a-go-go]

I would also like to add another two NICs for different networks?

any ideas how to configure this VM?

e

[einstein-a-go-go]

is this applicance now dead and un-supported?

I've been using successfully, but I've just noticed that one of the "pool" ntp servers is defunct, and hence the NTP server Ref time is bad.

it's never pulled down the list of 13+ ntp servers as listed, maybe it's because it's sat behind a proxy, and does not have direct access to the internet (where this does work well.

it's always worked well up until now, DNS and syncing has worked, it just doesn't pull the list of NTP servers, anyone know how to modify it with a static set!

[James_Bucknall]

Halon330,

When you're looking at the NTP peer Status on tty12 you should be able to return to the Main Menu console by pressing .

Re DNS: You'll either need to point the NTP to your Gateway router if it picks up and present the DNS provided by your ISP to the LAN. Alternatively configure the NTP to point to a public DNS on the internet. Check Firewall ports too!

Regards,

James

[rootdude]

James -

I'm having the same problem as the above poster - the VM states that it cannot resolve the DNS for the pool.ntp.org servers - I have it pointed to three internal DNS servers and have issued NSLOOKUPs against all of the internal DNS servers for pool.ntp.org as well as other ntp.org servers with success from another host on the network. UTC time seems right - but when I establish EST as my timezone, the resulting time is not correct either (these two problems are probably related).

The VM is useless at this point for me at least - I'm looking into ways of rolling my own if I can't get some fixes for these problems. Hate to discard this server - I like the idea - but the implementation is flawed.

Stac

[povergaard]

Has anyone any idea of how to use other NTP servers(internals) as source than the pool.ntp.org - eg. 10.134.1.254?

My appliance has no connection to the Internet.

[povergaard]

Halon330

Did you had any progress for this question?

"When booting the appliance never can resolve time servers to pull time from. Is there a way to configure what DNS servers the appliance uses. I tried adding two time servers manually by IP address, but I'm not sure that the appliance is using them"

[DIE-HARD]

The website to download this VM is gone.

Anyone still have this ovf laying arround?

This VM has worked great for me for some time now and wanted to roll it into some Lab networks I am building but I can't find what I did with the ovf!

[einstein-a-go-go]

PM me, and I'll send you a URL to download the original OVF

[mkizziah]

I'd also be interested in a copy of Appliance for your Lab. If it is still available

Cheers

MK