Re: CP VMDUMP for a mega server
CP VMDUMP 0-END TO OPERATNS That was it. ___ James Vincent Systems Engineering Consultant Nationwide Services Co., Technology Solutions Mainframe, z/VM and z/Linux Support One Nationwide Plaza 3-20-13 Columbus OH 43215-2220 U.S.A Voice: (614) 249-5547Fax: (614) 677-7681 mailto:[EMAIL PROTECTED] The IBM z/VM Operating System wrote on 02/27/2007 02:19:28 PM: > IBMVM@LISTSERV.UARK.EDU > > Jim, > > When you issued the VMDUMP command did you just enter #CP VMDUMP or #CP > VMDUMP 0-END FORMAT vmtype DSS TO dump_id ? > > If the second, what did you put in for "vmtype"? > > -Original Message- > From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On > Behalf Of Jim Vincent > Sent: Tuesday, January 02, 2007 3:58 PM > To: IBMVM@LISTSERV.UARK.EDU > Subject: CP VMDUMP for a mega server > > > We are doing a VMDUMP for a 2G linux server and it is taking FOREVER. > If this was a production server we would be picking daggers out of > places I can't mention here. > > Paging is not too bad on the system (300/sec) and we set the share and > reserved for the guest up to maybe help, but nothing seems to get it > moving any. It is doing about 1000 spool records every 15-20 seconds > and we guestimated it would be about 524288 records; after about an hour > it is less than half done!! > > Is this to be expected? Are we to expect VMDUMP to take that long for > these larger guests or do we have a situation we need to dig into? > > ___ > James Vincent > Systems Engineering Consultant > Nationwide Services Co., Technology Solutions > Mainframe, z/VM and z/Linux Support > One Nationwide Plaza 3-20-13 > Columbus OH 43215-2220 U.S.A > Voice: (614) 249-5547Fax: (614) 677-7681 > mailto:[EMAIL PROTECTED] > > > If you are not an intended recipient of this e-mail, please notify > the sender, delete it and do not read, act upon, print, disclose, > copy, retain or redistribute it. Click here for important additional > terms relating to this e-mail. http://www.ml.com/email_terms/ >
Re: CP VMDUMP for a mega server
Jim, Rick had asked me about this and I didn't think it was normal behaviour. I'm on the road, but have asked someone to look into this. Could you please open a PMR so we can track this? Thanks, Bit
Re: CP VMDUMP for a mega server
Jim, When you issued the VMDUMP command did you just enter #CP VMDUMP or #CP VMDUMP 0-END FORMAT vmtype DSS TO dump_id ? If the second, what did you put in for "vmtype"? -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Jim Vincent Sent: Tuesday, January 02, 2007 3:58 PM To: IBMVM@LISTSERV.UARK.EDU Subject: CP VMDUMP for a mega server We are doing a VMDUMP for a 2G linux server and it is taking FOREVER. If this was a production server we would be picking daggers out of places I can't mention here. Paging is not too bad on the system (300/sec) and we set the share and reserved for the guest up to maybe help, but nothing seems to get it moving any. It is doing about 1000 spool records every 15-20 seconds and we guestimated it would be about 524288 records; after about an hour it is less than half done!! Is this to be expected? Are we to expect VMDUMP to take that long for these larger guests or do we have a situation we need to dig into? ___ James Vincent Systems Engineering Consultant Nationwide Services Co., Technology Solutions Mainframe, z/VM and z/Linux Support One Nationwide Plaza 3-20-13 Columbus OH 43215-2220 U.S.A Voice: (614) 249-5547Fax: (614) 677-7681 mailto:[EMAIL PROTECTED] If you are not an intended recipient of this e-mail, please notify the sender, delete it and do not read, act upon, print, disclose, copy, retain or redistribute it. Click here for important additional terms relating to this e-mail. http://www.ml.com/email_terms/
Re: CP VMDUMP for a mega server
On 1/2/07, Jim Vincent <[EMAIL PROTECTED]> wrote: Is this to be expected? Are we to expect VMDUMP to take that long for these larger guests or do we have a situation we need to dig into? If you have detailed performance data from the event (or can afford to try again) it would be my pleasure to review that. Since you are paging, much of your virtual machine may be out on disk when you issue the vmdump. While your paging subsystem would allow for multiple paths, it may be that VMDUMP does only one at a time. -- Rob van der Heij Velocity Software, Inc http://velocitysoftware.com/
{SPAM?} RE: [IBMVM] CP VMDUMP for a mega server
OK, I tried it here (#CP VMDUMP 0-END on a 2G server). Took 15:27 minutes. Better than hours - but still seems long to wait when someone is screaming... Records ended up being 487788. This was z/VM 5.2 RSU 0602, z9-109 2 engines, 24G memory (20 central, 4xstor), heavily loaded with 80+ servers running - paging to DASD at approx 1000/sec. 2 virtual CPUs on the guest. 2 mod 9's for spool (IBM DS8300). Here's a SWAG- could all those spool writes with the little 4K blocks be overwhelming a certain vendors DASD write cache? How many i/o per sec are you getting off the spool volumes? Marcy Cortes This message may contain confidential and/or privileged information. If you are not the addressee or authorized to receive this for the addressee, you must not use, copy, disclose, or take any action based on this message or any information herein. If you have received this message in error, please advise the sender immediately by reply e-mail and delete this message. Thank you for your cooperation." -Original Message- From: The IBM z/VM Operating System [mailto:[EMAIL PROTECTED] On Behalf Of Jim Vincent Sent: Tuesday, January 02, 2007 12:58 To: IBMVM@LISTSERV.UARK.EDU Subject: [IBMVM] CP VMDUMP for a mega server We are doing a VMDUMP for a 2G linux server and it is taking FOREVER. If this was a production server we would be picking daggers out of places I can't mention here. Paging is not too bad on the system (300/sec) and we set the share and reserved for the guest up to maybe help, but nothing seems to get it moving any. It is doing about 1000 spool records every 15-20 seconds and we guestimated it would be about 524288 records; after about an hour it is less than half done!! Is this to be expected? Are we to expect VMDUMP to take that long for these larger guests or do we have a situation we need to dig into? ___ James Vincent Systems Engineering Consultant Nationwide Services Co., Technology Solutions Mainframe, z/VM and z/Linux Support One Nationwide Plaza 3-20-13 Columbus OH 43215-2220 U.S.A Voice: (614) 249-5547Fax: (614) 677-7681 mailto:[EMAIL PROTECTED]
CP VMDUMP for a mega server
We are doing a VMDUMP for a 2G linux server and it is taking FOREVER. If this was a production server we would be picking daggers out of places I can't mention here. Paging is not too bad on the system (300/sec) and we set the share and reserved for the guest up to maybe help, but nothing seems to get it moving any. It is doing about 1000 spool records every 15-20 seconds and we guestimated it would be about 524288 records; after about an hour it is less than half done!! Is this to be expected? Are we to expect VMDUMP to take that long for these larger guests or do we have a situation we need to dig into? ___ James Vincent Systems Engineering Consultant Nationwide Services Co., Technology Solutions Mainframe, z/VM and z/Linux Support One Nationwide Plaza 3-20-13 Columbus OH 43215-2220 U.S.A Voice: (614) 249-5547Fax: (614) 677-7681 mailto:[EMAIL PROTECTED]