toasters

toasters@lists.teaparty.net

13520 discussions

Problem mounting NFS from a Fedora Core 6 machine
by Gustavo Chaves 22 Nov '06

22 Nov '06

Hello, We have a F760 running Data Ontap 6.4.5. (Pretty old stuff, I know.) The filer has six network interfaces, each one connected to a different network. Recently we started experimenting with a Linux Fedora Core 6 machine. When we try to mount some directory at the filer from this machine, almost always the process hang. A very few times the process succeeds. Investigating the issue using tcpdump I discovered that in the process of mouting, when the filer's portmapper answers to the Linux box, sometimes its answers come from a different network interface and the Linux box doesn't like that. See below the history of an unsuccessful mount. The Linux box (10.202.40.66) tries to mount a directory from filer (10.202.48.109) with the following command and hangs until I kill it: mount filer:/vol/vol0/unix/config /mnt Note that on packet 111 the linux box sends a UDP packet to the filer's portmapper asking for the port number of the MOUNT service. The filer responds with packet 113, which comes from another of its interfaces (10.202.40.23). This response is rejected by the linux box with the ICMP packet number 114. If I try to mount specifying the interface 10.202.40.23, which is in the same network as the client, the mount succeeds. It seems that the portmapper response always tries to go back through the interface in the same network as the client. The problem is that we have some linux clients on networks on which there is no filer interface connected. In these cases it seems to be anyone's guess through which interface the portmapper response will come. We looked for differences in the TCP stack from Fedora Core 5 to 6, because until Core 5 this problem didn't arise, but we couldn't find anything relevant. Since the filer works ok with other Linux and Solaris machines, it seems that something must have been changed in the Linux kernel that made it more strict. We suspected of some firewall rules, but there is no iptables rules configured in the linux box. Has anybody seen this before? Any clue? Thank you. Gustavo. ---------------------------------------------------------------------- No. Time Source Destination Protocol Info 94 1.036676 10.202.40.66 10.202.48.109 TCP 33614 > sunrpc [SYN] Seq=0 Len=0 MSS=1460 TSV=16336999 TSER=0 WS=6 95 1.038301 10.202.48.109 10.202.40.66 TCP sunrpc > 33614 [SYN, ACK] Seq=0 Ack=1 Win=8760 Len=0 MSS=1460 WS=0 TSV=242744087 TSER=16336999 96 1.038319 10.202.40.66 10.202.48.109 TCP 33614 > sunrpc [ACK] Seq=1 Ack=1 Win=5888 Len=0 TSV=16336999 TSER=242744087 97 1.038385 10.202.40.66 10.202.48.109 Portmap V2 GETPORT Call (Reply In 98) NFS(100003) V:3 TCP 98 1.039907 10.202.48.109 10.202.40.66 Portmap V2 GETPORT Reply (Call In 97) Port:2049 99 1.039914 10.202.40.66 10.202.48.109 TCP 33614 > sunrpc [ACK] Seq=61 Ack=33 Win=5888 Len=0 TSV=16337000 TSER=242744087 100 1.039939 10.202.40.66 10.202.48.109 TCP 33614 > sunrpc [FIN, ACK] Seq=61 Ack=33 Win=5888 Len=0 TSV=16337000 TSER=242744087 101 1.039966 10.202.40.66 10.202.48.109 TCP 34612 > nfs [SYN] Seq=0 Len=0 MSS=1460 TSV=16337000 TSER=0 WS=6 102 1.040592 10.202.48.109 10.202.40.66 TCP sunrpc > 33614 [ACK] Seq=33 Ack=62 Win=8760 Len=0 TSV=242744087 TSER=16337000 103 1.040669 10.202.48.109 10.202.40.66 TCP nfs > 34612 [SYN, ACK] Seq=0 Ack=1 Win=26280 Len=0 MSS=1460 WS=0 TSV=242744087 TSER=16337000 104 1.040677 10.202.40.66 10.202.48.109 TCP 34612 > nfs [ACK] Seq=1 Ack=1 Win=5888 Len=0 TSV=16337000 TSER=242744087 105 1.040722 10.202.40.66 10.202.48.109 TCP [TCP segment of a reassembled PDU] 106 1.040849 10.202.48.109 10.202.40.66 TCP sunrpc > 33614 [FIN, ACK] Seq=33 Ack=62 Win=8760 Len=0 TSV=242744087 TSER=16337000 107 1.040856 10.202.40.66 10.202.48.109 TCP 33614 > sunrpc [ACK] Seq=62 Ack=34 Win=5888 Len=0 TSV=16337000 TSER=242744087 108 1.041776 10.202.48.109 10.202.40.66 NFS V3 NULL Reply (Call In 105) 109 1.041782 10.202.40.66 10.202.48.109 TCP 34612 > nfs [ACK] Seq=45 Ack=33 Win=5888 Len=0 TSV=16337000 TSER=242744087 110 1.041801 10.202.40.66 10.202.48.109 TCP 34612 > nfs [FIN, ACK] Seq=45 Ack=33 Win=5888 Len=0 TSV=16337000 TSER=242744087 111 1.041847 10.202.40.66 10.202.48.109 Portmap V2 GETPORT Call (Reply In 113) MOUNT(100005) V:3 UDP 112 1.042834 10.202.48.109 10.202.40.66 TCP nfs > 34612 [ACK] Seq=33 Ack=46 Win=26280 Len=0 TSV=242744087 TSER=16337000 113 1.042842 10.202.40.23 10.202.40.66 Portmap V2 GETPORT Reply (Call In 111) Port:4046 114 1.042851 10.202.40.66 10.202.40.23 ICMP Destination unreachable (Port unreachable) 115 1.042854 10.202.48.109 10.202.40.66 TCP nfs > 34612 [FIN, ACK] Seq=33 Ack=46 Win=26280 Len=0 TSV=242744087 TSER=16337000 116 1.042857 10.202.40.66 10.202.48.109 TCP 34612 > nfs [ACK] Seq=46 Ack=34 Win=5888 Len=0 TSV=16337000 TSER=242744087

1 0

RE: iSCSI SAN Queries
by Christopher Mende 21 Nov '06

21 Nov '06

I can't comment much on the other vendors' most current offerings, but iSCSI is anything but bolt-on for the filers. NetApp has the lion's share of iSCSI deployments worldwide. FAS270 (including clusters) running iSCSI configurations supporting Exchange, SQL, and all varieties of other applications are going in all over the place. With a FAS270c, tiered storage (mixing FC & SATA) is possible, with certain conditions. All secondary shelves are either SATA OR FC. From the system configuration guide: Caution 2: While both types of disk shelves can connect to the filer, if there is an existing DiskShelf14mk2 FC connected to the FAS270/FAS270c, then any additional expansion shelf must be a DiskShelf14mk2 FC. If however, you do not have an expansion shelf connected to the FAS270/FAS270c, then you can choose to connect to a loop of DiskShelf14mk2 AT or a loop of DiskShelf14mk2 FC, but not both. The FAS270/FAS270c does not support both types of disk shelves on the same loop. ________________________________ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Raj Patel Sent: Wednesday, 22 November 2006 8:40 a.m. To: toasters(a)mathworks.com Subject: iSCSI SAN Queries Hi, We're in the preliminary scoping phase for a low/mid-range iSCSI SAN solution. I was wondering if anyone had any positive or negative real-world experiences with - NetApp 270c HDS AMS200 EMC CX320 At the moment the NetApp is the most familiar to me as I have used one of their NAS boxes before and the simulator provides a pretty good indication of how it works. However the iSCSI seems a bit of a 'bolt-on' and its not clear if it will handle tiered storage as well as the other vendors (then again does it matter?). The HDS & EMC are unknown quantities (other than what I can glean from the web). Any feedback concerning ease of use, expansion, licensing, snapshot mechanism (the EMC seems clunky from their literature but I don't know if that's the case in operation). Also does anyone have any iSCSI 'gotchas' ? Is a TOE one of those 'nice to have but not really necessary' things on a modern server or should it be factored into the solution ? Cheers, Raj. The information contained in this email is privileged and confidential and intended for the addressee only. If you are not the intended recipient, you are asked to respect that confidentiality and not disclose, copy or make use of its contents. If received in error you are asked to destroy this email and contact the sender immediately. Your assistance is appreciated.

1 0

IBM Branded NetApps
by Raj Patel 19 Nov '06

19 Nov '06

Hi all, A few quick queries - Has anyone had any experience with the IBM Branded NetApp's ? What does the N5200 compare with in the NetApp family ? How does service/support differ from NetApp ? Finally why would anyone go with IBM for a NetApp rather than NetApp direct (is it purely a question of vendor service quality or is there a significant price difference) ? Thanks in advance, Raj.

7 8

RE: Measuring Filer Performance
by Willeke, Jochen 16 Nov '06

16 Nov '06

Hi, i agree that latency is one of the main factors. But sometimes i wonder how to measure the latency the best way. One way i know is to use "tethereal -i [interface] -z rpc,rtt,100003,3" Which gives me an overview of the rtt which is seen by the client. But i am not so sure what to use to see the output from the filer. I know that statit gives me some kind of rtt for each disk. But where do you get a good overall-response-time for your filers from? Those counters shown by Blake (...stat show...) are very good, but i think they only show what the head sees. What is a reasonable value to be added as an "head-overhead". For example: - I run tethereal from a direct-attached client and see a read-rtt of 12ms - I run stat show -i 1 nfsv3:nfs:nfsv3_read_latency and see 3ms - I run a ping from the direct-attached client and get a rtt of 2ms - So i would think that the filer-overhead is ~7ms Do you think this is correct way to go? Best Regards Jochen -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Glenn Walker Sent: Thursday, November 16, 2006 4:25 AM To: Blake Golliher; George, Andrew Cc: Rohit; toasters(a)mathworks.com Subject: RE: Measuring Filer Performance I'd agree with the latency standpoint. It's a good metric to determine when the users will start to bitch. The latency might be hidden by the application, but it depends on how well it's written (single threaded vs multi threaded\low concurrency vs high concurrency). Typically, this is all a function of CPU\DISK (and sometimes memory) load. As a result, those are good to watch, and they are a direct result of the overall workload (CIFS\NFS\iSCSI IOPS). The hard part is determining what\why the workload tanks. Most times, users are only going to feel impact if there is a long running event that generates high latency - however, depending on workload, a single spike that lasts only moments (a second or two) can be devastating: this is where knowing your environment comes in handy. Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Blake Golliher Sent: Wednesday, November 15, 2006 7:08 PM To: George, Andrew Cc: Rohit; toasters(a)mathworks.com Subject: Re: Measuring Filer Performance Latency is primary for what most of what my filers are doing. Usually watched via nfs_hist (being sure to -z the stats), or using stats all wrapped up in a shell script. rsh $FILER 'priv set -q diag ; stats show -i $INTER -n $SAMPLES volume:$VOL:avg_latency volume:$VOL:total_ops volume:$VOL:read_latency volume:$VOL:read_ops volume:$VOL:write_latency volume:$VOL:write_ops volume:$VOL:other_latency volume:$VOL:other_ops volume:$VOL:nfs_read_latency volume:$VOL:nfs_read_ops volume:$VOL:nfs_write_latency volume:$VOL:nfs_write_ops volume:$VOL:nfs_other_latency volume:$VOL:nfs_other_ops' nfsv3:nfs:nfsv3_avg_op_latency nfsv3:nfs:nfsv3_op_percent nfsv3:nfs:nfsv3_write_ops nfsv3:nfs:nfsv3_write_latency nfsv3:nfs:nfsv3_read_ops nfsv3:nfs:nfsv3_read_latency nfsv3:nfs:nfsv3_ops' It's good to note that while stats show averages, nfs_hist shows a histogram, and can spot the occasional very high latency operation. Good to watch. I just poked around, there is a cifs component for stats, and it shows some interesting metrics the cifs users might want to watch. filer01*> stats show cifsdomain cifs cifs:cifs:cifs_ops:2658/s cifs:cifs:cifs_op_count: GetAttr: 1296 Read: 989 Write: 0 Lock: 1 Open/Close: 23 Directory: 0 Other: 335 cifs:cifs:cifs_op_pct: GetAttr: 48% Read: 37% Write: 0% Lock: 0% Open/Close: 0% Directory: 0% Other: 12% cifs:cifs:cifs_latency:3.27ms cifsdomain:10.0.0.3:netlogon_latency:9.50ms cifsdomain:10.0.0.3:netlogon_latency_base:6 cifsdomain:10.0.0.3:lsa_latency:0ms cifsdomain:10.0.0.3:lsa_latency_base:0 cifsdomain:10.0.0.3:samr_latency:0ms cifsdomain:10.0.0.3:samr_latency_base:0 cifsdomain:10.0.0.2:netlogon_latency:2.00ms cifsdomain:10.0.0.2:netlogon_latency_base:1 cifsdomain:10.0.0.2:lsa_latency:0ms cifsdomain:10.0.0.2:lsa_latency_base:0 cifsdomain:10.0.0.2:samr_latency:0ms cifsdomain:10.0.0.2:samr_latency_base:0 cifsdomain:10.0.0.1:netlogon_latency:0ms cifsdomain:10.0.0.1:netlogon_latency_base:0 cifsdomain:10.0.0.1:lsa_latency:0ms cifsdomain:10.0.0.1:lsa_latency_base:0 cifsdomain:10.0.0.1:samr_latency:0ms cifsdomain:10.0.0.1:samr_latency_base:0 There's also smb_hist. That's from a 960 running 6.5.6PsomethingDsomething.... -Blake On 11/15/06, George, Andrew <georgea(a)anz.com> wrote: > We only use our filers for CIFS (so this may be completely inapplicable > to you) > > 1. Cache age! Mainly because when I have seen filers that are worked to > death, that's the value that tends to react to it (though CP there are a > lot of secondary metrics to support that). > > 2. As mentioned by someone else...the magic metric is how badly the > users are bitching. That being a little "rubbery" we also check cache > age, cp type, disk utilisation, cpu usage, Ops/sec and network traffic > to determine if anything's bottlenecking. > > -----Original Message----- > From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] > On Behalf Of Rohit > Sent: Wednesday, 15 November 2006 9:59 PM > To: toasters(a)mathworks.com > Subject: Measuring Filer Performance > > Hi Folks > > I had some open questions and i hope you'll help me answer them based on > your experience. > > 1. If you just had a single metric to determine filer performance what > would it be ? > Would it be CPU usage, throughput, latency, IOPS..or anything else ? > > 2. At your place, what metrics do you use to determine if a filer is > still good enough not to worry about ? > > thanks > rohit > > ------------------------------------------------------------------------ > -- > Refinance & Home Equity Loan Quotes > Low credit score? Old Merchants can help. Get a quote in 2 easy steps! > http://tagline.bidsystem.com/fc/BgLEQfI7zNZsu5fZHhXyj0UMjSNzY6ZfSaWC/ > > > > > "This e-mail and any attachments to it (the "Communication") is, unless otherwise stated, confidential, may contain copyright material and is for the use only of the intended recipient. If you receive the Communication in error, please notify the sender immediately by return e-mail, delete the Communication and the return e-mail, and do not read, copy, retransmit or otherwise deal with it. Any views expressed in the Communication are those of the individual sender only, unless expressly stated to be those of Australia and New Zealand Banking Group Limited ABN 11 005 357 522, or any of its related entities including ANZ National Bank Limited (together "ANZ"). ANZ does not accept liability in connection with the integrity of or errors in the Communication, computer virus, data corruption, interference or delay arising from or in respect of the Communication." >

1 0

RE: Measuring Filer Performance
by Glenn Walker 16 Nov '06

16 Nov '06

I'd agree with the latency standpoint. It's a good metric to determine when the users will start to bitch. The latency might be hidden by the application, but it depends on how well it's written (single threaded vs multi threaded\low concurrency vs high concurrency). Typically, this is all a function of CPU\DISK (and sometimes memory) load. As a result, those are good to watch, and they are a direct result of the overall workload (CIFS\NFS\iSCSI IOPS). The hard part is determining what\why the workload tanks. Most times, users are only going to feel impact if there is a long running event that generates high latency - however, depending on workload, a single spike that lasts only moments (a second or two) can be devastating: this is where knowing your environment comes in handy. Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Blake Golliher Sent: Wednesday, November 15, 2006 7:08 PM To: George, Andrew Cc: Rohit; toasters(a)mathworks.com Subject: Re: Measuring Filer Performance Latency is primary for what most of what my filers are doing. Usually watched via nfs_hist (being sure to -z the stats), or using stats all wrapped up in a shell script. rsh $FILER 'priv set -q diag ; stats show -i $INTER -n $SAMPLES volume:$VOL:avg_latency volume:$VOL:total_ops volume:$VOL:read_latency volume:$VOL:read_ops volume:$VOL:write_latency volume:$VOL:write_ops volume:$VOL:other_latency volume:$VOL:other_ops volume:$VOL:nfs_read_latency volume:$VOL:nfs_read_ops volume:$VOL:nfs_write_latency volume:$VOL:nfs_write_ops volume:$VOL:nfs_other_latency volume:$VOL:nfs_other_ops' nfsv3:nfs:nfsv3_avg_op_latency nfsv3:nfs:nfsv3_op_percent nfsv3:nfs:nfsv3_write_ops nfsv3:nfs:nfsv3_write_latency nfsv3:nfs:nfsv3_read_ops nfsv3:nfs:nfsv3_read_latency nfsv3:nfs:nfsv3_ops' It's good to note that while stats show averages, nfs_hist shows a histogram, and can spot the occasional very high latency operation. Good to watch. I just poked around, there is a cifs component for stats, and it shows some interesting metrics the cifs users might want to watch. filer01*> stats show cifsdomain cifs cifs:cifs:cifs_ops:2658/s cifs:cifs:cifs_op_count: GetAttr: 1296 Read: 989 Write: 0 Lock: 1 Open/Close: 23 Directory: 0 Other: 335 cifs:cifs:cifs_op_pct: GetAttr: 48% Read: 37% Write: 0% Lock: 0% Open/Close: 0% Directory: 0% Other: 12% cifs:cifs:cifs_latency:3.27ms cifsdomain:10.0.0.3:netlogon_latency:9.50ms cifsdomain:10.0.0.3:netlogon_latency_base:6 cifsdomain:10.0.0.3:lsa_latency:0ms cifsdomain:10.0.0.3:lsa_latency_base:0 cifsdomain:10.0.0.3:samr_latency:0ms cifsdomain:10.0.0.3:samr_latency_base:0 cifsdomain:10.0.0.2:netlogon_latency:2.00ms cifsdomain:10.0.0.2:netlogon_latency_base:1 cifsdomain:10.0.0.2:lsa_latency:0ms cifsdomain:10.0.0.2:lsa_latency_base:0 cifsdomain:10.0.0.2:samr_latency:0ms cifsdomain:10.0.0.2:samr_latency_base:0 cifsdomain:10.0.0.1:netlogon_latency:0ms cifsdomain:10.0.0.1:netlogon_latency_base:0 cifsdomain:10.0.0.1:lsa_latency:0ms cifsdomain:10.0.0.1:lsa_latency_base:0 cifsdomain:10.0.0.1:samr_latency:0ms cifsdomain:10.0.0.1:samr_latency_base:0 There's also smb_hist. That's from a 960 running 6.5.6PsomethingDsomething.... -Blake On 11/15/06, George, Andrew <georgea(a)anz.com> wrote: > We only use our filers for CIFS (so this may be completely inapplicable > to you) > > 1. Cache age! Mainly because when I have seen filers that are worked to > death, that's the value that tends to react to it (though CP there are a > lot of secondary metrics to support that). > > 2. As mentioned by someone else...the magic metric is how badly the > users are bitching. That being a little "rubbery" we also check cache > age, cp type, disk utilisation, cpu usage, Ops/sec and network traffic > to determine if anything's bottlenecking. > > -----Original Message----- > From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] > On Behalf Of Rohit > Sent: Wednesday, 15 November 2006 9:59 PM > To: toasters(a)mathworks.com > Subject: Measuring Filer Performance > > Hi Folks > > I had some open questions and i hope you'll help me answer them based on > your experience. > > 1. If you just had a single metric to determine filer performance what > would it be ? > Would it be CPU usage, throughput, latency, IOPS..or anything else ? > > 2. At your place, what metrics do you use to determine if a filer is > still good enough not to worry about ? > > thanks > rohit > > ------------------------------------------------------------------------ > -- > Refinance & Home Equity Loan Quotes > Low credit score? Old Merchants can help. Get a quote in 2 easy steps! > http://tagline.bidsystem.com/fc/BgLEQfI7zNZsu5fZHhXyj0UMjSNzY6ZfSaWC/ > > > > > "This e-mail and any attachments to it (the "Communication") is, unless otherwise stated, confidential, may contain copyright material and is for the use only of the intended recipient. If you receive the Communication in error, please notify the sender immediately by return e-mail, delete the Communication and the return e-mail, and do not read, copy, retransmit or otherwise deal with it. Any views expressed in the Communication are those of the individual sender only, unless expressly stated to be those of Australia and New Zealand Banking Group Limited ABN 11 005 357 522, or any of its related entities including ANZ National Bank Limited (together "ANZ"). ANZ does not accept liability in connection with the integrity of or errors in the Communication, computer virus, data corruption, interference or delay arising from or in respect of the Communication." >

2 1

Measuring Filer Performance
by Rohit 16 Nov '06

16 Nov '06

Hi Folks I had some open questions and i hope you'll help me answer them based on your experience. 1. If you just had a single metric to determine filer performance what would it be ? Would it be CPU usage, throughput, latency, IOPS..or anything else ? 2. At your place, what metrics do you use to determine if a filer is still good enough not to worry about ? thanks rohit -------------------------------------------------------------------------- Refinance & Home Equity Loan Quotes Low credit score? Old Merchants can help. Get a quote in 2 easy steps! http://tagline.bidsystem.com/fc/BgLEQfI7zNZsu5fZHhXyj0UMjSNzY6ZfSaWC/

3 2

RE: Measuring Filer Performance
by George, Andrew 16 Nov '06

16 Nov '06

We only use our filers for CIFS (so this may be completely inapplicable to you) 1. Cache age! Mainly because when I have seen filers that are worked to death, that's the value that tends to react to it (though CP there are a lot of secondary metrics to support that). 2. As mentioned by someone else...the magic metric is how badly the users are bitching. That being a little "rubbery" we also check cache age, cp type, disk utilisation, cpu usage, Ops/sec and network traffic to determine if anything's bottlenecking. -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Rohit Sent: Wednesday, 15 November 2006 9:59 PM To: toasters(a)mathworks.com Subject: Measuring Filer Performance Hi Folks I had some open questions and i hope you'll help me answer them based on your experience. 1. If you just had a single metric to determine filer performance what would it be ? Would it be CPU usage, throughput, latency, IOPS..or anything else ? 2. At your place, what metrics do you use to determine if a filer is still good enough not to worry about ? thanks rohit ------------------------------------------------------------------------ -- Refinance & Home Equity Loan Quotes Low credit score? Old Merchants can help. Get a quote in 2 easy steps! http://tagline.bidsystem.com/fc/BgLEQfI7zNZsu5fZHhXyj0UMjSNzY6ZfSaWC/ "This e-mail and any attachments to it (the "Communication") is, unless otherwise stated, confidential, may contain copyright material and is for the use only of the intended recipient. If you receive the Communication in error, please notify the sender immediately by return e-mail, delete the Communication and the return e-mail, and do not read, copy, retransmit or otherwise deal with it. Any views expressed in the Communication are those of the individual sender only, unless expressly stated to be those of Australia and New Zealand Banking Group Limited ABN 11 005 357 522, or any of its related entities including ANZ National Bank Limited (together "ANZ"). ANZ does not accept liability in connection with the integrity of or errors in the Communication, computer virus, data corruption, interference or delay arising from or in respect of the Communication."

2 1

Re: Measuring Filer Performance
by Reinoud Reynders 15 Nov '06

15 Nov '06

Hi Rohit, My view is that your application performance is the most important. This is what our users see and feel. The problem is that for every application, the impact of the storage on the performance is different. For our Databases, responstime is the key. So for those applications we measure the latency. For our PACS (digital radiology) is Mb/s more important. So we try to measure the application performance and the key infrastructure elements of that application (like latency). On filer level, we monitor the CPU to give a general idee of the load of the filer. Best regards, Reinoud Reynders IT manager Infrastructure & Operations Leuven University Hospital -----Original Message----- From: owner-toasters(a)mathworks.com <owner-toasters(a)mathworks.com> To: toasters(a)mathworks.com <toasters(a)mathworks.com> Sent: Wed Nov 15 11:58:39 2006 Subject: Measuring Filer Performance Hi Folks I had some open questions and i hope you'll help me answer them based on your experience. 1. If you just had a single metric to determine filer performance what would it be ? Would it be CPU usage, throughput, latency, IOPS..or anything else ? 2. At your place, what metrics do you use to determine if a filer is still good enough not to worry about ? thanks rohit -------------------------------------------------------------------------- Refinance & Home Equity Loan Quotes Low credit score? Old Merchants can help. Get a quote in 2 easy steps! http://tagline.bidsystem.com/fc/BgLEQfI7zNZsu5fZHhXyj0UMjSNzY6ZfSaWC/

1 0

RE: IBM Branded NetApps
by Julio Calderon 15 Nov '06

15 Nov '06

I've been in the same spot having to deal with a vendorA when vendorA hd to get support from VendorB and I could never get quick and advance help from anyone! Not a fun place to be when you need help. Good luck -----Original Message----- From: Zayas, Russell [mailto:rzayas@qualcomm.com] Sent: Tue Nov 14 16:19:46 2006 To: Toasters; Raj Patel; toasters(a)mathworks.com Subject: RE: IBM Branded NetApps Does that mean that support is geared to handle tough issues? Or do they pipe you into NetApp support at that point? Curious, -Russ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Toasters Sent: Tuesday, November 14, 2006 12:41 PM To: Raj Patel; toasters(a)mathworks.com Subject: RE: IBM Branded NetApps We just went with a few IBM boxes here. We have 40 or so Netapp boxes. The determining factor for us was the price. They undercut by well over 100k on what we bought. The support structure is not the same as Netapp. If you have never experienced IBM support before then I suggest you do before you decided to go the way of IBM/NAS. If your like us and don't need the "what is a snapmirror" type of support and just call when you need parts or the box is panicing then you may want to consider them. E- ________________________________ From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Raj Patel Sent: Tuesday, November 14, 2006 2:03 PM To: toasters(a)mathworks.com Subject: IBM Branded NetApps Hi all, A few quick queries - Has anyone had any experience with the IBM Branded NetApp's ? What does the N5200 compare with in the NetApp family ? How does service/support differ from NetApp ? Finally why would anyone go with IBM for a NetApp rather than NetApp direct (is it purely a question of vendor service quality or is there a significant price difference) ? Thanks in advance, Raj.

1 0

RE: SnapMirror Bandwidth restrictions
by Davies,Matt 14 Nov '06

14 Nov '06

Usally less than 1 second, at least a couple of times during the snapmirror job It is a Volume snapmirror Ontap 7.1 -----Original Message----- From: Glenn Walker [mailto:ggwalker@mindspring.com] Sent: 14 November 2006 13:32 To: Davies,Matt; toasters(a)mathworks.com Subject: RE: SnapMirror Bandwidth restrictions Hmm - I honestly have never noticed that problem on any of our equipment... How long does the spike last? What version of ONTAP? VSM or QSM? Glenn -----Original Message----- From: Davies,Matt [mailto:MDAVIES@generalatlantic.com] Sent: Tuesday, November 14, 2006 7:58 AM To: Glenn Walker; toasters(a)mathworks.com Subject: RE: SnapMirror Bandwidth restrictions Glenn, The snapmirror.conf file is showing 1000 as the limit. Which should equate to just under 8Mb/s Overall the rate does average out for the snapmirror jobs to be below 1000KB/s There was only one snapmirror running at the time of the test I carried out to confirm the rates. Matt -----Original Message----- From: Glenn Walker [mailto:ggwalker@mindspring.com] Sent: 14 November 2006 12:47 To: Davies,Matt; toasters(a)mathworks.com Subject: RE: SnapMirror Bandwidth restrictions I assume that the snapmirror.conf file is showing 1000 as the limit? This is in KB (kilobyte), not kb(kilobit). Are you sure that there aren't multiple mirrors running simultaneously? Glenn -----Original Message----- From: owner-toasters(a)mathworks.com [mailto:owner-toasters@mathworks.com] On Behalf Of Davies,Matt Sent: Tuesday, November 14, 2006 5:40 AM To: toasters(a)mathworks.com Subject: SnapMirror Bandwidth restrictions Hi, We have noticed that when a snapmirror runs between our filers with a maximum transfer rate of 1000 KB/s, we are seeing peaks of traffic of around 24Mb/s. This has been confirmed using ethereal, that it is the snapmirror traffic. We have noticed that these peaks are causing excessive amount of dropped packets on the wan interface of the routers between all the site. we have tried a number of difference queuing strategies and bandwidth shaping, but the router cannot buffer all the packets during the peaks and starts dropping them. As we also use these links for both voice and video, we are seeing high jitter rates during these peaks, and need to do something about it. We have a mixture of FAS250's and FAS270's all are showing the same problem Has anyone else experienced this ? Or does anyone have any ideas. Cheers Matt

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

toasters