Hi David
Thanks for your advice.
I thus setup an NFS I have mounted on each cluster of my datacenter (same absolute path on each node). See the print screens below.
I had a close look the the property file, but I do not see how I can point my shared file-system.
Moreover, VMmark/Auction fails with RC=1 which refer to an API issue (see the messages bellow).
Any idea of what is going wrong? I am running out of ideas.
Thanks.
20190522-11:02:14 Info Process: Client0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variable doesnot exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:02:14 Info Client0 : Get Guest Info did not run correctly
20190522-11:02:14 Info Process: AuctionLB0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variable does not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:02:14 Info AuctionLB0 : Get Guest Info did not run correctly
20190522-11:02:14 Info Process: localhost : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variable does not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:02:14 Info localhost : Get Guest Info did not run correctly
20190522-11:02:14 Info GuestInfo Results : AuctionLB0 :: Failed : Errors 0 : ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:02:14 Info GuestInfo Results : Client0 :: Failed : Errors 0 : ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:02:14 Info GuestInfo Results : localhost :: Invalid : Errors 1 : ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:03:14 Info Process: AuctionWebA0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variabledoes not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:03:14 Info AuctionWebA0 : Get Guest Info did not run correctly
20190522-11:03:15 Info GuestInfo Results : AuctionWebA0 :: Failed : Errors 0 :ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:04:15 Info Process: AuctionWebB0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variabledoes not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:04:15 Info AuctionWebB0 : Get Guest Info did not run correctly
20190522-11:04:15 Info GuestInfo Results : AuctionWebB0 :: Failed : Errors 0 :ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:05:15 Info Process: AuctionAppA0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variabledoes not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:05:15 Info AuctionAppA0 : Get Guest Info did not run correctly
20190522-11:05:15 Info GuestInfo Results : AuctionAppA0 :: Failed : Errors 0 :ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:06:15 Info Process: AuctionAppB0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variabledoes not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:06:15 Info AuctionAppB0 : Get Guest Info did not run correctly
20190522-11:06:16 Info GuestInfo Results : AuctionAppB0 :: Failed : Errors 0 :ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:07:16 Info Process: AuctionDB0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variable does not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:07:16 Info AuctionDB0 : Get Guest Info did not run correctly
20190522-11:07:16 Info GuestInfo Results : AuctionDB0 :: Failed : Errors 0 : ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:08:16 Info Process: AuctionMSQ0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variable does not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:08:16 Info AuctionMSQ0 : Get Guest Info did not run correctly
20190522-11:08:16 Info GuestInfo Results : AuctionMSQ0 :: Failed : Errors 0 :ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:09:16 Info Process: AuctionNoSQL0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variabledoes not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190522-11:09:16 Info AuctionNoSQL0 : Get Guest Info did not run correctly
20190522-11:09:17 Info GuestInfo Results : AuctionNoSQL0 :: Failed : Errors 0: ErrorMsgs : Warnings 0 : WarningMsgs
20190522-11:09:47 Info Guest Info failures for the following machines: 10 machines : Client0 AuctionLB0 localhost AuctionWebA0 AuctionWebB0 AuctionAppA0 AuctionAppB0 AuctionDB0 AuctionMSQ0AuctionNoSQL0 Error Messages Client0= :AuctionLB0= :localhost= :AuctionWebA0= :AuctionWebB0= :AuctionAppA0=:AuctionAppB0= :AuctionDB0= :AuctionMSQ0= :AuctionNoSQL0= :
Hi mathieugont
Sorry you're still having issues. A few things I want to point out:
All of that aside, though, there seems to be a fundamental networking problem causing the "Guest Info failures". If you SSH into your PrimeClient VM, are you able to ping the tile0 VMs (i.e. 'ping Standby0, 'ping DS3WebA0')?
Hi mathieugont
Sorry you're still having issues. A few things I want to point out:
All of that aside, though, there seems to be a fundamental networking problem causing the "Guest Info failures". If you SSH into your PrimeClient VM, are you able to ping the tile0 VMs (i.e. 'ping Standby0, 'ping DS3WebA0')?
Hi David
I followed your recommendations grouping all my nodes within a single cluster (4 hardware servers grouped in ClusterESXi_BDW), except one (in ClusterESXi_Mgr) hosting the VMware vCenter Server Appliance (named photon), and a personal VM used to access vSphere client remotely (internal config). See the print screen below.
In ClusterESXi_BDW, all my VMs are hosted by the node 172.22.197.109, while my Client0 and my Prime Client are on the node 172.22.193.115.
From my Prime Client, I can ping and passwordless ssh all the VMs.
About a well-performing scores without high-performing storage (i.e. SSD arrays), the aim is to make it work first. Score and improvement will be treated in a second hand. For the moment, I need to run weathervane only, that is why in my property files, I set:
WorkloadList = AuctionLB,AuctionWebA,AuctionWebB,AuctionAppA,AuctionAppB,AuctionDB,AuctionMSQ,AuctionNoSQL
InfrastructureList =
But as you can see in the zip file attached to this message, comment "InfrastructureList =" to get the default does not solve the problem.
Quick additional comment
Hi mathieugont
I checked with the technical lead for VMmark, he recommended rebooting your PrimeClient VM and double-checking that there aren't any duplicate IP addresses in your network. It looks like something aborted the Guest Info collection on the PrimeClient -- these kinds of messages are very unusual:
20190531-18:07:34 Info Process: localhost : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variable does not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190531-18:07:34 Info localhost : Get Guest Info did not run correctly
20190531-18:07:34 Info Process: AuctionLB0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variable does not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190531-18:07:34 Info AuctionLB0 : Get Guest Info did not run correctly
20190531-18:07:34 Info Process: Client0 : Get Guest Info failed to start/complete. Returned: RC = 1, STAFResult = This variable doesnot exist: VMMARK/AuctionNoSQL/SKIPRESTORE
20190531-18:07:34 Info Client0 : Get Guest Info did not run correctly
Also -- if you want to just run Weathervane, that workload is an open-source project on github:
GitHub - vmware/weathervane: Virtual-Infrastructure and Cloud Performance Benchmark
Weathervane itself has a run harness that might be easier to use than trying to run VMmark without the other workloads, but it's up to you.
Thanks David
Did you saw the same message appeared whatever the VM:
Returned: RC = 1, STAFResult = This variabledoes not exist: VMMARK/AuctionNoSQL/SKIPRESTORE
RC=1 refers to API issues. Not a good sign 😉
But seems AuctionNoSQL is the issue.
Any comment on that?
David
I am trying to run weathervane out of VMmark like you advised to me.
I followed the procedures described in section 3 and 4 of the manual. I needed to tune the scripts, but it the end, it does not work with the following message:
Wed Jun 5 14:46:29 2019: Weathervane Version 1.0.21
Wed Jun 5 14:46:29 2019: Command-line parameters:
Wed Jun 5 14:46:29 2019: Run Configuration has 1 workloads.
Wed Jun 5 14:46:29 2019: Workload 1 has 1 workload-driver nodes
Wed Jun 5 14:46:29 2019: Workload 1 has 1 application instances.
Wed Jun 5 14:46:29 2019: Workload 1, Application Instance 1 configuration:
Wed Jun 5 14:46:29 2019: 1 ConfigurationManagers
Wed Jun 5 14:46:29 2019: 0 ElasticityServices
Wed Jun 5 14:46:29 2019: 1 CoordinationServers
Wed Jun 5 14:46:29 2019: 0 LbServers
Wed Jun 5 14:46:29 2019: 1 WebServers
Wed Jun 5 14:46:29 2019: 1 DbServers
Wed Jun 5 14:46:29 2019: 1 NosqlServers
Wed Jun 5 14:46:29 2019: 0 FileServers
Wed Jun 5 14:46:29 2019: 1 MsgServers
Wed Jun 5 14:46:29 2019: 1 AppServers
Wed Jun 5 14:46:29 2019: maxDuration = 7200
Wed Jun 5 14:46:29 2019: Running Weathervane with Single-Run RunManager using Full-Run RunProcedure.
Wed Jun 5 14:46:29 2019: Single-Run RunManager starting run.
Wed Jun 5 14:46:30 2019: Stopping running services and cleaning up old log and stats files.
Wed Jun 5 14:46:48 2019: Preparing data for use in current run.
Wed Jun 5 14:46:48 2019: Configuring and starting data services for appInstance 1 of workload 1.
Wed Jun 5 14:47:40 2019: Backup is not available at the proper scale for 1000 users for appInstance 1 of workload 1. Loading data.
Wed Jun 5 14:48:29 2019: Workload 1, appInstance 1: Loading data for a maximum of 1000 users
Wed Jun 5 14:49:00 2019: Can't find dbloader pid for workload 1, appInstance 1
Wed Jun 5 14:49:29 2019: Could not properly load or prepare data for run 13. Exiting.
Any idea of what is going wrong?
Hi mathieugont, I'm working on this and will get you an update tomorrow.
Thanks a lot
Take your time: Monday is off here 😉
Hi, I noticed you are using a non-default name for VMmark3.properties. Could you try reverting your file VMmark3.tile0.weathervane-auction.properties back to the default name VMmark3.properties (and following the instructions in the User's Guide to support that under "Starting a VMmark Run Through The GUI"). Also please comment out the line DS3DB/DS3DBsize = 100.
Also, could you please zip the results file from your most recent run and attach it to this thread.
Thank you,
Rebecca