Hi,

After a few years running a windows 2003 server as guest on vwware server 1 and ubuntu 7.04, yesterday failed to start up.

I see a blue start up screen with the message UNMOUNTABLE_BOOT_DEVICE,

Aftter googling a while I see It can be solved using chkdsk.

But I failed trying to start from safe mode, and after using the windows 2003 cd to use the recovery console I receive the next messege

"Setup did not find any hard disk drives installed in your computer"

I also use a windows 7 dvd to run recovery but it didn't show the partition.

Please somebody has any idea how to solve this?

Any help will be aprecciated

TIA

Jesus

It's not currently possible to use a USB device both in a VMware guest and in the host at the same time, correct?

Per my observations, the guest takes the device away from the host OS when it powers up (assuming all the USB stuff is set up for the guest, etc).

Obviously we can set up a network share of the filesystem to make it accessible, but just wondering if there is any feasible workaround for this.

Has anyone one gotten Microsoft Windows 3.51 Server running on VMWare Server?

If so, can you please explain how you did it?

I tryed using VMWare Converter, but it does not allow Windows 3.51 Server.

Thanks

All,

I've just set up VMWare Server 1.0.2 on a box running CentOS-64 4.4 (RHEL4 clone). I've configured Kerberos on the Host OS to allow local users to authenticate via their Kerberos password (i.e. "adduser johndoe" creates the user on the server but their password comes from the Kerberos server), but these users cannot log into the VMware Server Console. Full server-based users ("root" and a "vmware" user I created) can log in, but not the Kerb-Authenticated users.

I'm assuming my solution lies somewhere in the /etc/vmware/pam.d/vmware-authd file, but I haven't found mention of this anywhere in the documentation aside from the "VMware Server for Linux uses Pluggable Authentication Modules" blurb in the Server Admin guide.

Help?

\- John

Hi, I've installed VMW Server 1.0.4 b56528 on WinXP Pro SP2.

Freshly installed VMW Server and after resetting the computer (without even running VMW Console) I get an application event error.

here are 6 events that appear at once, every 5 seconds, flooding the event viewer.

1. Description:

The description for Event ID ( 100 ) in Source ( vmauthd ) cannot be found. The local computer may not have the necessary registry information or message DLL files to display messages from a remote computer. You may be able to use the /AUXSOURCE= flag to retrieve this description; see Help and Support for details. The following information is part of the event: Failed to read registry perf object Memory\Committed Bytes

2. Same description but: Failed to read registry perf object Memory\Pool Paged Bytes
3. Same description but: Failed to read registry perf object Memory\Pool Nonaged Bytes
4. Same description but: Failed to read registry perf object Memory\Pool Faults/sec
5  Same description but: Failed to read registry perf object Process\ID Process
6. Same description but: Failed to read registry perf object Memory\Cache Bytes

Already visited and performed actions explained here:

• http://communities.vmware.com/message/421234?jsessionid=a172f2b198dde4f342f01d2c67ae99fb

• http://communities.vmware.com/message/1154#1154

• Re-entered a new serial.

• Installed Extensible Performance Counter List (exctrlst.exe) and checked ALL the items on the list.

• Stopped/Started (and restarted) all VMware services

• vmware-authd.exe service is Logged On as Local System Account.

• Disabled Antivirus and Firewall.

Stating the obviuos:  If I stop the VMware Authorization Service, the problem goes away; then of course VMware is disconnected and doesn't function.

Services and running processes are shown on the attached file.

No success.

Please help.

After spending what now amounts to 4 weeks dealing with this issue.  Get ready for a long post hopefully with lots of performance/throughput numbers.

The challenge:

Windows 2000 SP4 Guest

  >512MB RAM

  >2 CPU (tried with 1 in many test also, just prolonged the inevitable)

4GB OS vmdk (IDE)

  >Separate 12GB (tried 20GB also) persistent, pre-allocated SCSI drive, 2GB Files (tried as one file, no diff)  >Running IOMeter 2004.07.30 on the full 12GB drive, 100% Sequential 64k Writes (or Reads, depending on test)  Continuous test, watching "Results Since" at 1 sec update and getting a "mental" average watching for extremes that might be averaged out.

The hardware:

Supermicro X7DBE motherboard

2x Xeon Quad Core 2.33GHz E5345 (8 cores total)

8GB RAM

2 Raptor 74GB drives (mirrored for system)

6 73GB 15k SAS drives connected to Adapted 3805 SAS controller

Running in RAID5 (64k stripe, I think, maybe 128k, yes I know RAID10 is faster, but I'd prefer the space and as you'll see below, it should be plenty fast)

The goal: I want to run this server under a variant of linux, preferably Fedora Core 6 (FC6), CentOS 5, or Ubuntu 6.06 LTS (Preferably a 2.6.18 kernel, since that's required for LM sensors to work with that board (Actually .20 is required, but the module backports and compiles to .18 just fine, FYI)  I want it on linux for several reasons: licensing, flexibility, DRBD, higher performance*, etc.

The problem: Windows seems to have way better disk performance in guests than linux, not good...

VMWare on Windows Server 2003 R2 x64

Host Disk Performance (IOMeter 2004.07.30), same settings as guest above:

220MB/s read, 150MB/s write (fyi 1MB request size bumps up to 315MB/s & 195MB/s)  This is sustained, the initial burst (and this goes for all the host tests, and some of the guest test as well) can peak up to 2GB/s, some run at 500-600MB/s for a several seconds before dropping down.  FYI 4k 100% rnd read 250iops read 700iops write (??!?)

Windows version of dd writes 12GB file to disk at 258MB/s

2000 Guest under Win2k3

Read:  180-247 Avg 230MB/s

Write: 110-122 Avg 122MB/s

Very good, that's what I was expecting, on to the linux tests:

Tried FC5 x64 yum updated with latest kernel 2.6.20, BAD MOVE, as well documented in the forum, 2.6.19 and greater the vm (and machine) will crash/lock up.  Need to stay with 2.6.18 or lower (though 2.6.16 and 2.6.17 have other problems I've read of on the forums)

CentOS 4.4 x64 (yum updated)

Host Disk Performance (iometer slaves running under linux, manager running on windows)

226MB/s read, 170MB/s write (better than windows here, all linuxes performed similarly in host testing)

2000 Guest under CentOS 4.4 x64 2.6.9-55? kernel

Read:  Avg 240MB/s (But unstable)

Write: Avg 38MB/s ?!!!!!!!?!???! Thus starts the problems (very unstable)

Sisoft Sandra 1117XP 2CPU 14298MIPS/13986MFLOPS (running in guest)

Decided to run with this anyway, but when I converted as server over it had weird "pausing" problems, evidenced by the clock VISIBLY stopping and loosing time (a several seconds per MINUTE!)  Moved host over to windows server so I could reformat and continue testing.

FC6 x64 (NO kernel update, stick with stock 2.6.18)

Host Disk Performance (same as CentOS), still good

2000 Guest under FC6 2.6.18 kernel

Read:  Avg 50MB/s (But unstable)

Write: Avg 20MB/s (very unstable goes to 0 for several seconds some times, CentOS did same thing, vmware logs "Command WRITE(10) took N.NNN seconds" up to 15 seconds)

Sisoft Sandra 1117XP 2CPU 20804MIPS/14271MFLOPS (running in guest), newer kernels have better cpu performance in guests.

Better performance in Sandra, feels a little snappier too, but disk performance is still poor, see conclusions below.

UBuntu 6.06.1LTS x64 2.6

Host 208MB/s read, 140MB/s Write (used iometer 2006)

2000 Guest:

Read:  68MB/s

Write: 30MB/s

Tried CentOS 5 x64, CentOS 4.5 x32 (Note, saw 8GB), CentOS 5 x32 (4GB) also same.

Finally tested internal drive, has a better ratio to the raw disk performance:

Internal Raptor drive 80MB/s read and write (sustained)

2000 Guest under CentOS 5 x32 with internal drive

Read:  Avg 43MB/s (Stable)

Write: 0-45 Avg 38MB/s Same bad performance

Troubleshooting sucesses and failures:

Noticed Cyclic relationship watching iostat -m 1 with iometer running in the guest, the transfer rate would go up in the guest, while host would stay low, then the guest would drop way down and host would flush it out to disk (with a notably higher rate 40 for guest vs 60 for host), then the cycle would start over.  5-6 second cycle at best after some tweaking.  I played around with tweaking /proc/sys/vm/dirty_writeback_centisecs and setting it to 60 or with best result setting it to 0 (tested both on array and internal drive)  It served to fix that 14 second delay (now only a second or two at worst), but it was still cyclic.  Played with other settings in /proc/sys/vm to no major effect.

Messed with hdparm on the block devices, helped with read performance on host, minimal effect on guest (my standard is hdparm -a 8192 /dev/sxx)  16384 hurt performance.

Most host os tests were done (with the exception of a few iometer tests) with dd if=/dev/zero of=somefileonarraytotest bs=1M count=20000 (or 12000) always returned 230-260MB/s  because of the large ram anything less than 8GB would return up to 2.5GB/s

MemTrimRate=0

sched.mem.pshare.enable = "FALSE"

mainMem.useNamedFile = "FALSE"

Tried above with no major effect on disk performance (actually seemed to slow interface down a bit perhaps)

Tried using latest driver from adaptec (aacraid), no effect, still bad.

Was initially doing tests with XFS, but changed to ext3 to use CentOS versions.  Made no difference either way though, host performance and guest performance still same.

My conclusion (but not solution) is that because of memory mapping or buffering or both (or ???) that there is a buffer that is being allowed to fill up, then drain while the guest waits for it to drain, then start over again.  The bad read performance is very curious on the newer kernels, WHY!  Most of all WHY IS WINDOWS PERFORMING SO MUCH BETTER.  I can understand a moderate difference like if the guest had 150MB/s reads and 100MB/s writes, but not 0-40MB/s writes !!

Also note that using noatime on a partition caused VM Server Console not to be able to create a new virtual disk !?!?? (Under CentOS 4.4)  Turned off and was fine.  Also suspend/resume from disk on shutdown/startup did not work under CentOS 4.4, worked fine elsewhere.

There were lots more things I tried that I can't remember/didn't write down.  Scoured the net for answers and this is all I could find. 

Need your help.

(Learned a lot about linux over the last several weeks, from boot problems to all kinds of stuff, regardless almost all these versions of linux have installed relatively painlessly on this brand new latest hardware and the OSes have been very stable)

Hi,

This question has probably been asked a million times but i can't find documentation on it. I have WinXP SP3 which is running VMserver and on the VM i'm using WinXP SP3. My question is how do i create a share or a share drive so i can transfer documents quickly between the VM and my host OS.

Thank you

Hi, I was wondering if it's possible for the vmrun executable that comes with VIX 1.6.2 to work with VMware Server 1.08?  All examples I've seen online show syntax for using vmrun with a remote VMware Server 2.0 machine.

If so, is there any configuration needed beyond installing the VIX API?

Thanks,

Joe

PS - sorry if this is a dupe, I can't seem to find a forum search function

I have a problem with a guest machine running w2k. on this machine there is an application that uses apache, it all installs ok but just refuses to work. the apache tomcat logs are full of the following

istenning on the wrong port. Failed errno = 61

  http://jk_ajp_common.c (955): Error connecting to the Tomcat process.

  http://jk_ajp_common.c (1309): sending request to tomcat failed in send loop. err=1

  http://jk_connect.c (233): jk_open_socket, connect() failed errno = 61

  http://jk_ajp_common.c (676): Error connecting to tomcat. Tomcat is probably not started or is listenning on the wrong port. Failed errno = 61

  http://jk_ajp_common.c (955): Error connecting to the Tomcat process.

  http://jk_ajp_common.c (1309): sending request to tomcat failed in send loop. err=2

  http://jk_ajp_common.c (1318): Error connecting to tomcat. Tomcat is probably not started or is listening on the wrong port. worker=ajp13 failed errno = 61

could this be a port clash with vmware??

hello,

I'm running debian4 as host and a win2k3 server web edition as guest, and I receive this error message in my guest, every time I reboot the host:

could not get interface flags for vmnet1

...this inevitably leads to the network interface not being connected in the guest OS (the VM)

To solve the problem I have to either "connect" the NIC manually after start-up in the server console - or reboot the guest in the host web interface.

any ideas on how to make the guest start later so that the vmnet1 interface has time to "come up" before the guest starts?

Continue Starting Virtual Machines After - 6 minutes (in the Server settings in the web interface) has no apparent effect...it still starts it "instantly" without any 6 minute delay.

or start the guests after my network interfaces have been started?

Thanks a lot in advance for any insights!

Hi,

I can not copy any file from other (real) machines on any of my Virtuel machines (VmWare 1.0.3 build 44356) The error message is :

File creation error - The specified network name is no longer available.

The OS of virtuel machines and physical real machines is Win 2003 server.

Any idea ?

Many thanks.

hi,this is what i have done:1) acronis image from metal2) created a vm over vmware converter (source: acronis image) -->everything fine, the vm comes up and after a minute the cpu-usage rise to 100%(process: system)thanks

Hello!

I have Vista Sp1 host with VMWare Server 1.0.4 installed. After installation I cannot burn DVD-R, but DVD-RW burns well. CDBurnerXP shows error message : "Could not write to Disk, LBA: xxx Size: yyyy" just after burn process start. I've tried reinstalling VMWare but it didn't work out.

Could someone help with this issue?

Running VMWare server 1.0.5.  Host and guest are XP Professional 32 bit.  I'm trying to shrink the disk on this VM and I cannot get the button to appear to shrink it.  The disk is NOT independent mode (I tried it both ways.)  The shrink tab says that shrink is disabled because of ...  I checked and I don't think any of these apply to my disk.  It is not a preallocated disk.  The same server can shrink at least one other VM on the same disk.  I have attached the list of files, the log and my VMX file.

Any help you could give would be appreciated.

Thanks.

Hi,

I was just wondering if running Mac OSX 10.5 Leopard in VMWare Server 1.8 for Windows was a supported, or at least working, comnfiguration ?

Thanks for the feedback,

John Smith

guys

I know zero about VMware so bare with me. I've been trying to create a vm of our Blackberry server. I have VM converter (3.0.3 build 89816) installed locally on the  BES.  when i run the convert it gets to 72% and then fails,looking at the loogs it looks like this is the error:

CloneTask::task: Image processing task has failed with PlatformError fault: 734004

I've attached the converter logs to this post

Hello,

I have vmvare-server 1.0.8, running on Debian Etch 2.6.18 kernel and since upgrade  virtual hosts (debian Etch again) to version 2.6.24-etchnhalf.1-686, 2.6.24-etchnhalf.1-686-bigmem, 2.6.26-bpo.2-686-bigmem or 2.6.26-bpo.2-686, virtual OSes silently freeze after some time.

In case od 2.6.24 kernel after 1-3hours, in case of 2.6.26 kernels, after 1-2 days.

Any idea?

My windows guest (only have two Windows 2000 SP4 servers) the clock is very very very slow ( the windows clock runs one second every 6-7 real seconds).

My host server is a cuad core:

Intel(R) Core(TM)2 Quad CPU    Q6600  @ 2.40GHz

8Gb Ram

vmware server 1.0.8

Debian Lenny 5.0

Customized kernel: 2.6.28.7.090413 #1 SMP

The real clock runs well, I check my .config file and see that has  CONFIG_HZ_250=y

(the server runs very slow after of some time) but the sql server runs well (I think).

The config of my guest windows are:

config.version = "8"

virtualHW.version = "4"

numvcpus = "2"

scsi0.present = "TRUE"

memsize = "1024"

scsi0:0.present = "TRUE"

scsi0:0.fileName = "Windows_2000_Server.vmdk"

ide1:0.present = "FALSE"

ide1:0.fileName = "auto detect"

ide1:0.deviceType = "atapi-cdrom"

floppy0.fileName = "/dev/fd0"

Ethernet0.present = "TRUE"

displayName = "Windows_2000_Srv_PIFA2"

guestOS = "win2000serv"

autostop = "softpoweroff"

priority.grabbed = "normal"

priority.ungrabbed = "normal"

timeTracker.periodicStats="TRUE"

timeTracker.statsIntercal="10"

ide1:0.autodetect = "TRUE"

floppy0.present = "FALSE"

scsi0:0.redo = ""

ide1:0.startConnected = "TRUE"

ethernet0.startConnected = "TRUE"

ethernet0.addressType = "generated"

uuid.location = "56 4d fd 0f 9e 67 62 5e-d6 3c 37 1e 48 e3 1b 59"

uuid.bios = "56 4d fd 0f 9e 67 62 5e-d6 3c 37 1e 48 e3 1b 59"

ethernet0.generatedAddress = "00:0c:29:e3:1b:59"

ethernet0.generatedAddressOffset = "0"

tools.syncTime = "TRUE"

parallel0.present = "TRUE"

parallel0.fileName = "/dev/parport0"

parallel0.bidirectional = "FALSE"

1. This is because I have a netHASP key that runs well

ide0:0.present = "TRUE"

ide0:0.fileName = "/dev/scd0"

ide0:0.deviceType = "atapi-cdrom"

workingDir = "."

ide0:0.startConnected = "FALSE"

autostart = "poweron"

I use in same configuration same kernel and same hardware in a  windows XP guest and the clock runs well

I have ACPI configurated in same machines.

I try.

host.cpukHz = 2400075000

host.noTSC = TRUE

ptsc.noTSC = TRUE

Nothink happeds.

My disable acpi to solve this problem, and why in windows xp don’t happends??

Thanks for the answers!!!!

Need to know as to what and all config backups we need to take from an ESX machine (config info of VM, Virtual switch and all networking info). And how to restore it after upgrading with new ESX. And also if there are any existing API's available (any language) to perform this task.

Hi,

Can any body guide me to edit the size of Linux vm using VM Converter , OR by any other way

Regards & Thanks

Yogesh S