Message-Id: 7.0.1.0.2.20060312064321.01e8f418@free.fr X-Mailer: QUALCOMM Windows Eudora Version 7.0.1.0 Date: Sun, 12 Mar 2006 06:50:16 +0100 To: Alan Sparks asparks@doublesparks.net, Toasters List toasters@mathworks.com From: Jean-Jacques ARNOUX arnouxjj@free.fr Subject: Re: "D" CP types on FAS3020 In-Reply-To: 4413A4EC.50905@doublesparks.net References: 4413A4EC.50905@doublesparks.net Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="=====================_1004222343==.ALT"
--=====================_1004222343==.ALT Content-Type: text/plain; charset="us-ascii"; format=flowed
Hi Alan,
i meen that you reach this bug. I have the same on one of my customer site. And i have a huge number of this message.
Sat Mar 11 23:04:11 CET [stk-hfm1: APIPool01:debug]: Removing too-old iterator file: /etc/.zapi/7537813126.index Sat Mar 11 23:04:11 CET [stk-hfm1: APIPool01:debug]: Removing too-old iterator file: /etc/.zapi/7537813126.value Sat Mar 11 23:04:11 CET [stk-hfm1: APIPool01:debug]: Removing too-old iterator file: /etc/.zapi/7537813126.next Sat Mar 11 23:35:49 CET [stk-hfm1: APIPool04:debug]: Removing too-old iterator file: /etc/.zapi/7539708117.index Sat Mar 11 23:35:49 CET [stk-hfm1: APIPool04:debug]: Removing too-old iterator file: /etc/.zapi/7539708117.value Sat Mar 11 23:35:49 CET [stk-hfm1: APIPool04:debug]: Removing too-old iterator file: /etc/.zapi/7539708117.next Sun Mar 12 00:00:00 CET [stk-hfm1: kern.uptime.filer:info]: 12:00am up 87 days, 7:48 1447100364 NFS ops, 7295964634 CIFS ops, 4291 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops
on others systems i have these messages before upgrading to 7.0.3 or 7.1
as you see on one other system nothing Sat Mar 11 20:00:00 CET [grizzly: kern.uptime.filer:info]: 8:00pm up 49 days, 55 mins, 21659709149 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops Sat Mar 11 21:00:00 CET [grizzly: kern.uptime.filer:info]: 9:00pm up 49 days, 1:55 21697185323 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops Sat Mar 11 22:00:00 CET [grizzly: kern.uptime.filer:info]: 10:00pm up 49 days, 2:55 21732428903 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops Sat Mar 11 23:00:00 CET [grizzly: kern.uptime.filer:info]: 11:00pm up 49 days, 3:55 21761084021 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops Sun Mar 12 00:00:00 CET [grizzly: kern.uptime.filer:info]: 12:00am up 49 days, 4:55 21782379473 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops
but i have no performance issues on both sites.
Regards
At 05:34 12/03/2006, Alan Sparks wrote:
Not been getting a clear answer from Support on this, maybe someone here has some good advice or insight...
Our FAS3020C filers often display CP types of "D" in a sysstat listing. I get that this is a CP due to "log datavecs." Trying to understand if this is a resource problem, or at least indicative of a performance issue: CPU Total Net kB/s Disk kB/s Tape kB/s Cache Cache CP CP Disk ops/s in out read write read write age hit time ty util 17% 1166 12787 488 793 15683 0 0 >60 100% 55% D 12% 17% 937 13021 454 720 14376 0 0 >60 100% 49% D 10% 18% 1321 13167 538 697 14219 0 0 >60 100% 49% D 10%
One thing I got from Support is that it might be indicative of bug 145410 (must actively reclaim resources). At their suggestion, I've updated to DOT 7.0.3 (update from 7.0.2), but still see them a lot of the time.
Is this pretty normal for the FAS3000 series? Thanks in advance.
-- Alan Sparks, UNIX/Linux Systems Integration and Administration asparks@doublesparks.net
--=====================_1004222343==.ALT Content-Type: text/html; charset="us-ascii"
<html> <body> Hi Alan,<br><br> i meen that you reach this bug. I have the same on one of my customer site. And i have a huge number of this message.<br><br> <font face="Courier New, Courier">Sat Mar 11 23:04:11 CET [stk-hfm1: APIPool01:debug]: Removing too-old iterator file: /etc/.zapi/7537813126.index<br> Sat Mar 11 23:04:11 CET [stk-hfm1: APIPool01:debug]: Removing too-old iterator file: /etc/.zapi/7537813126.value<br> Sat Mar 11 23:04:11 CET [stk-hfm1: APIPool01:debug]: Removing too-old iterator file: /etc/.zapi/7537813126.next<br> Sat Mar 11 23:35:49 CET [stk-hfm1: APIPool04:debug]: Removing too-old iterator file: /etc/.zapi/7539708117.index<br> Sat Mar 11 23:35:49 CET [stk-hfm1: APIPool04:debug]: Removing too-old iterator file: /etc/.zapi/7539708117.value<br> Sat Mar 11 23:35:49 CET [stk-hfm1: APIPool04:debug]: Removing too-old iterator file: /etc/.zapi/7539708117.next<br> Sun Mar 12 00:00:00 CET [stk-hfm1: kern.uptime.filer:info]: 12:00am up 87 days, 7:48 1447100364 NFS ops, 7295964634 CIFS ops, 4291 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops<br><br> </font>on others systems i have these messages before upgrading to 7.0.3 or 7.1<br><br> as you see on one other system nothing<br> <font face="Courier New, Courier">Sat Mar 11 20:00:00 CET [grizzly: kern.uptime.filer:info]: 8:00pm up 49 days, 55 mins, 21659709149 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops<br> Sat Mar 11 21:00:00 CET [grizzly: kern.uptime.filer:info]: 9:00pm up 49 days, 1:55 21697185323 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops<br> Sat Mar 11 22:00:00 CET [grizzly: kern.uptime.filer:info]: 10:00pm up 49 days, 2:55 21732428903 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops<br> Sat Mar 11 23:00:00 CET [grizzly: kern.uptime.filer:info]: 11:00pm up 49 days, 3:55 21761084021 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops<br> Sun Mar 12 00:00:00 CET [grizzly: kern.uptime.filer:info]: 12:00am up 49 days, 4:55 21782379473 NFS ops, 0 CIFS ops, 1051 HTTP ops, 0 DAFS ops, 0 FCP ops, 0 iSCSI ops<br><br> </font>but i have no performance issues on both sites.<br><br> Regards<br><br> At 05:34 12/03/2006, Alan Sparks wrote:<br> <blockquote type=cite class=cite cite="">Not been getting a clear answer from Support on this, maybe someone here has some good advice or insight...<br><br> Our FAS3020C filers often display CP types of "D" in a sysstat listing. <br> I get that this is a CP due to "log datavecs." Trying to understand if this is a resource problem, or at least indicative of a performance issue:<br> CPU Total Net kB/s Disk kB/s Tape kB/s Cache Cache CP CP Disk<br> ops/s in out read write read write age hit time ty util<br> 17% 1166 12787 488 793 15683 0 0 >60 100% 55% D 12%<br> 17% 937 13021 454 720 14376 0 0 >60 100% 49% D 10%<br> 18% 1321 13167 538 697 14219 0 0 >60 100% 49% D 10%<br><br> One thing I got from Support is that it might be indicative of bug 145410 (must actively reclaim resources). At their suggestion, I've updated to DOT 7.0.3 (update from 7.0.2), but still see them a lot of the time.<br><br> Is this pretty normal for the FAS3000 series? Thanks in advance.<br><br> -- <br> Alan Sparks, UNIX/Linux Systems Integration and Administration<br> <asparks@doublesparks.net><br><br> </blockquote></body> <br> </html>
--=====================_1004222343==.ALT--
-
arnouxjj＠free.fr