toasters March 2003

toasters@lists.teaparty.net

92 participants
102 discussions

WWW over NFS
by Robert Borowicz 19 Mar '03

19 Mar '03

Has anybody located their root website on a filer and mounted it via NFS to their webserver? I don't believe it would need "root=" permissions. I'm considering this possibility but am curious towards the Security considerations. IE: We'd have to open the NFS port through out Firewall to the DMZ to allow the NFS mount. How secure is the NFS port in OnTAP? Thanks -Bob

4 3

RE: Moving files but not .snapshots?
by Mohler, Jeff 18 Mar '03

18 Mar '03

No..it uses the builtin loopback/localhost IP pipe. Im not sure if theyve improved it, but IP performance over that used to be abysmal compared to GigE. -----Original Message----- From: Jeffrey.Burton(a)abbott.com [mailto:Jeffrey.Burton@abbott.com] Sent: Thursday, March 13, 2003 3:39 PM To: Art Hebert Cc: Jim Davis; toasters(a)mathworks.com Subject: RE: Moving files but not .snapshots? I don't know if the network card is in the path of an internal copy, but I'm fairly certain it does not. I copied 400M internally on my F880C in just under 2 minutes. I'm sure it depends heavily on the amount of fragmentation on the source's file system (my wafl scan measure_layout was 1.13, which is almost perfect). For some reason I can't find the ndmp throttle (to limit the CPU usage of ndmpcopy), I'm guessing that option does not exist, too bad. Below copied from NOW at: http://now.netapp.com/Knowledgebase/solutionarea.asp?id=ntapcs2471&resource= Thanks, Jeff ======================================== Beginning with Data ONTAP 6.2, ndmpcopy is built into the filer and does not require a seperate host to migrate data. See your respective Data ONTAP Data Protection Guide for more details: http://now.netapp.com/NOW/knowledge/docs/ontap/ontap_index.shtml Complete the following steps: 1. Obtain a current version of ndmpcopy from http://www.ndmp.org. 2. Install ndmpcopy on a host that has network access to the source and destination filers. 3. On both filers, enable NDMPD: a. Enter ndmpd on b. To confirm NDMP is running, enter ndmpd status. 4. Establish a "trust" relationship between the client and filer. a. Enter the hostname or interface name , i.e. , of the client and the destination filer in the source filer's /etc/hosts.equiv file. See the hosts.equiv manual page for additional options. 5. Execute the ndmpcopy on the client. The general syntax is ndmpcopy source destination Refer to ndmpcopy documentation for additional options. For other ways to migrate data, see the following: http://www.netapp.com/tech_library/3018.html ======================================== Art Hebert <art(a)arzoon.com> 03/13/2003 04:55 PM To: "'Jeffrey.Burton(a)abbott.com'" <Jeffrey.Burton(a)abbott.com>, Jim Davis <jdavis(a)cs.arizona.edu> cc: toasters(a)mathworks.com Subject: RE: Moving files but not .snapshots? I take it when copying files using ndmpcopy on a single source and destination filer the data never uses the network. This seems like it should be a lot faster, anybody have an idea how much faster? Also, which release was the first with ndmpcopy on board? Is there any other ways to copy just on the filer itself beside ndmpcopy? thanks art -----Original Message----- From: Jeffrey.Burton(a)abbott.com [mailto:Jeffrey.Burton@abbott.com] Sent: Thursday, March 13, 2003 12:19 PM To: Jim Davis Cc: toasters(a)mathworks.com Subject: Re: Moving files but not .snapshots? You can use ndmpcopy from the filerhead CLI, granted you have a new enough DOT version. This creates a snapshot then copy's all the data and ACL's from source to target. First make sure that the ndmp option is turned on: ndmpd.enable on Then below will work: ndmpcopy /vol/sourcevol /vol/destvol You can copy at any level, not just the vol level. If the source and dest's vols are on a seperate filer just preceed the dir with "filename:" so "ndmpcopy sourcefiler:/vol/sourcevol destfiler:/vol/destvol". After the copy you can remove the temp file created at the destination end which is used for incremental updates (I forgot the name but it will be a huge file with an obviously generated name). Man page and NOW has good ndmp info. Thanks, Jeff Jim Davis <jdavis(a)cs.arizona.edu> Sent by: owner-toasters(a)mathworks.com 03/13/2003 01:47 PM To: toasters(a)mathworks.com cc: Subject: Moving files but not .snapshots? I want to move two big directories from an old volume to a new volume, but just the files -- not the snapshot subdirectories. So far I've tried find . -type d -name '.snapshot' -prune -o -print | cpio -pdm /newvol from the adminhost, which works but isn't blazingly fast. Are there any other ways people can suggest?

2 1

Rancho HVD-LVD converter
by Todd Leuffgen 18 Mar '03

18 Mar '03

We have recently purchased a LVD SDLT 320 tape drive that we were hoping could plug directly into our F720 Filer. Problem is that filer has a HVD SCSI connection on the back of it and no ability to add a LVD SCSI card to it (at least that is what Netapp Support says). So I was wondering if anyone out there has tried a Rancho LVD to HVD SCSI converter in this type of application and if so what were the results. Thanks in advance, Todd Leuffgen

3 2

未承諾広告※緊急のお知らせ
by me465156 15 Mar '03

15 Mar '03

<MÒ> dq[LÐ cÆLÅÍ èÜ¹ñ §104-0061 sæâÀ8-19-3 æ2ECOr@3F TEL:03-3544-6222 FAX:03-3544-6218 ºLÌæ€ÈàÌÍŒ\[ÈÌÅA²ÓŸ³¢B íQÉíêœûÍÅñÌx@ÉíQÍ¯ðoµÄŸ³¢B ±ÌŒ\Í³·ÊÉçêÄéàÌÅ·BàeââsûÀà ÏŠÄM³êÄÜ·ªAÂ Œ\Å·B ±ÌdbÆe`wÍÐÌàÌÅ·ªAèÉgíêÄ¢éàÌÅA ÐÆÍÜÁœÖW èÜ¹ñB ±ÌgÑÔÍu090-1281-3920ATU-KACF€m/ò/Od/ÃªvÆlÌàÌÆvíêÜ·B {¶Œ\àeF yä¿Êmz så}äAvµÜ·t ±ÌxÍßÉ Èœlªgp³êœdbñü©çÚ±³êœ A_gTCgp¿àÉÂ¢Ä ^cÆÒæè¢[p¿àÉÖ·éÂ ÷nðó¯ €ª¢[p¿àÌñûìÆð ãs³¹ÄžÉÈèÜµœÌÅäA³¹Äž«Ü·B »ÝÍºLÉLÚÌp¿àª¢[ÆÈÁÄÜ·ÌÅ x¹QàšæÑñûãsè¿àÜßÄ QQUúijßãRÜÅÌUèÝð äx¥¢úÀÆµÄºLÉLÚÌwèûÀÜÅ äüàµÄžæ€äè¢\µã°Ü·B všx¥¢àzFQORPO~ @@@@^cÆÒFsNmds @@¢[p¿àFPQTOO~ @@@x¹QàF@RPQO~ @ñûãsè¿F@SUXO~ yUæûÀz ======================= @ eâsðäp·éê@@@@@ @@@ÝžÙâsFVhŒûxX @@@@ûÀÔFjSRRROSV @@@@ûÀŒ`FV~Y C`E @ XÖÇðäp·éê @@@@Ê LFPOPVO @@@@Ê ÔFVPQWSRTP @@@@Ê Œ`FV~Y C`E ===================================== UèÝÌÛÉÍKžñû®Ôð ŒOÌOÉäüÍµÄº³¢B ¬â©ÉäüàµÄž¯È¢êÍ enæÌÂ ÖAÆÒšæÑÖA±Ö o^îñšæÑÂlîñðó¯n·ÌÅ ÅIIÉWàêåSõðä©îÈÇÉ Kâð³¹Äž«Ü·B »ÌÛÉÍãLÌvx¥zÉ ðÊïÆlïðÁZµÄ ñPO{Ì¿³¹ÄžêªäÀ¢Ü·ÌÅ šYêÈKžäüàµÄº³¢B xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx @@@@@ ij®Æ^ã\FŽ êY @@@Â îñÇVXe @@@sdkF03-3544-6222 @@@e`wF03-3544-6218 @@@@gÑF090-1281-3920 @@@@SFº pr@ Œ\ÌàeASÒÌŒOÍÏŠÄMµÄ¢Ü·B ê®êà±ÌèÌŒ\É²Óº³¢B

1 0

Resize existing volume
by Patrick Lee 14 Mar '03

14 Mar '03

I have a Filer 720. Is there any way to resize an existing volume short of re-creating it? Originally the volume started with all 18 GB drives. Overtime, as disks failed, they were replaced with 36 GB drives. So now we have some 36 GB drives that are only half used. See below: RAID Disk HA.ID HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ----- ------------ ---- -------------- -------------- parity 8.1 8 0 1 FC:A 34500/70656000 35003/71687368 data 8.4 8 0 4 FC:A 17000/34816000 34732/71132960 data 8.9 8 0 9 FC:A 17000/34816000 34732/71132960 data 8.3 8 0 3 FC:A 17000/34816000 17366/35566480 data 8.2 8 0 2 FC:A 17000/34816000 17366/35566480 data 8.0 8 0 0 FC:A 17000/34816000 17366/35566480 data 8.7 8 0 7 FC:A 34500/70656000 34732/71132960 Thanks, -- Patrick Lee <pat(a)patlee.org>

1 0

RE: backup permission problem on nfs share
by James Beal 14 Mar '03

14 Mar '03

1 0

backup permission problem on nfs share
by Danny G Smith 14 Mar '03

14 Mar '03

I am using Time Navigator to backup an nfs share on an f85 filer. I have given the linux bacup server "dilbert" root access via the exports file: /vol/volxtal/x -root=dilbert:omega,access=alpha:beta I have run an exportfs to make sure that everything is ok. This is an example of the log from tina: 4097|14|unix_end_read_obj_data|9|~|1|4|1047654637|2184|tina_backup|dilbert|root|xtal|dilbert|Unable to set time on "sfcheck.log", retcode = ERR_PERM|863| 4097|14|unix_end_read_obj_data|9|~|1|4|1047654639|2184|tina_backup|dilbert|root|xtal|dilbert|Unable to set time on "sfcheck_dcr.dat", retcode = ERR_PERM|863| thanks, Danny G Smith ----------------------------- Sr UNIX Systems Administrator Crystallography Dept Oklahoma Medical Research Foundation phone (405)271-8954 fax (405)271-7953 ========================================================= UNIX air conditioned environment, keep the windows closed

1 0

RE: mixing BCS and ZCS in the same volume
by Michael Schipp 14 Mar '03

14 Mar '03

Hello Steve, From the NOW site. Solution ID: ntapcs753 Title: Frequently asked questions on Zoned and Block Checksums - NetApp Filer In short is says that you can mix them but BCS will not be able to act as a spare (in you case you need an ZCS as a spare or will suffer degraded mode if a drive fails. Q. Can I add a BCS disk to an existing volume group which is currently using zoned checksums? A. Yes, they can all be added but cannot all be used as spares. With the BCS checksum format, 18 GB and 72 GB drives, can be added to a current ZCS volume as a spare in a ZCS or BCS raid group. However a 36 GB BCS drive cannot be used as a spare for a 36 GB ZCS dirve. Note also, that no Zoned Checksum disk can be used as a replacement for a BCS checksum formatted disk. Note the following exception: Q. Are the FC9 and DS14 34 GB disks identical in size and block capacity A. The 36 GB BCS formatted disk CANNOT be used as a replacement spare disk for the 36 GB Zoned Checksum disk because the 36 GB BCS disk right sizes to 34 GB and the 36 GB Zoned Checksum disk right sizes to 34.5 GB, and, therefore, does not have the same storage capacity. Q. Will I still get the performance boost from the BCS disk when it is part of a ZCS volume? A. No. Checksums are not stored in the extra 64 bits of the BCS (520B/sector) disk blocks; therefore, the volume still uses either ZCS or, for pre-Data ONTAP 6.0, no checksums at all. Q. Can a volume have more than one checksum style? A. The checksum style applies to the entire volume. You cannot have a volume with both checksum styles. Thank you Michael A Schipp ASI Solutions IT LAN/WAN Administrator Phone : 02 9384 8095 Mobile : 0402 907 928 E-Mail : mschipp(a)asi.com.au Visit us at http://www.asi.com.au -----Original Message----- From: steve(a)hq.newdream.net [mailto:steve@hq.newdream.net] Sent: Friday, 14 March 2003 1:12 PM To: toasters(a)mathworks.com Subject: mixing BCS and ZCS in the same volume Is it bad to mix BCS and ZCS drives in the same volume? A person from a reputable Network Appliance supply company told me that this will probably most definitely lead to data corruption. Please read below, I know its a lot of information but it should be useful for someone to give me a straight answer. before we just had one FC9 shelf of seven 36GB ZCS disks on our F740 filer slot 8: Fibre Channel Host Adapter 8 (Qlogic 2100 rev. 3, 32-bit, L-port, <UP>) 0: SEAGATE ST136403FC NA11 34.5GB 512B/sect 1: SEAGATE ST136403FC NA11 34.5GB 512B/sect 2: SEAGATE ST136403FC NA11 34.5GB 512B/sect 3: SEAGATE ST136403FC NA11 34.5GB 512B/sect 4: SEAGATE ST136403FC NA11 34.5GB 512B/sect 5: SEAGATE ST136403FC NA11 34.5GB 512B/sect 6: SEAGATE ST136403FC NA11 34.5GB 512B/sect now we have another FC9 shelf of 36GB BCS disks but the sizes don't match slot 8: Fibre Channel Host Adapter 8 (Qlogic 2100 rev. 3, 32-bit, L-port, <UP>) 0: SEAGATE ST136403FC NA11 34.5GB 512B/sect 1: SEAGATE ST136403FC NA11 34.5GB 512B/sect 2: SEAGATE ST136403FC NA11 34.5GB 512B/sect 3: SEAGATE ST136403FC NA11 34.5GB 512B/sect 4: SEAGATE ST136403FC NA11 34.5GB 512B/sect 5: SEAGATE ST136403FC NA11 34.5GB 512B/sect 6: SEAGATE ST136403FC NA11 34.5GB 512B/sect 8: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 9: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 10: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 11: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 12: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 13: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect Before we added the second shelf I had it setup very nicely with a boot volume of 2 disks, data volume of 4 disks (w/one spare) and raid group size=7. Volume boot (online, normal) (zoned checksums) Plex /boot/plex0 (online, normal, active) RAID group /boot/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.0 8 0 0 FC:A 34500/70656000 34732/71132960 data 8.1 8 0 1 FC:A 34500/70656000 34732/71132960 Volume saggy (online, normal) (zoned checksums) Plex /saggy/plex0 (online, normal, active) RAID group /saggy/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.2 8 0 2 FC:A 34500/70656000 34732/71132960 data 8.3 8 0 3 FC:A 34500/70656000 34732/71132960 data 8.4 8 0 4 FC:A 34500/70656000 34732/71132960 data 8.5 8 0 5 FC:A 34500/70656000 34732/71132960 Then a person with very little experience (my boss) added another shelf to the data volume but without first checking the types of disks he was adding, etc. He added NETAPP X231 (BCS) disks to this ZCS voume. These are block checksum disks of a different physical specs. Now it looks like this. Volume saggy (online, normal) (zoned checksums) Plex /saggy/plex0 (online, normal, active) RAID group /saggy/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.2 8 0 2 FC:A 34500/70656000 34732/71132960 data 8.3 8 0 3 FC:A 34500/70656000 34732/71132960 data 8.4 8 0 4 FC:A 34500/70656000 34732/71132960 data 8.5 8 0 5 FC:A 34500/70656000 34732/71132960 data 8.6 8 0 6 FC:A 34500/70656000 34732/71132960 data 8.8 8 1 0 FC:A 34000/69632000 34190/70022840 data 8.9 8 1 1 FC:A 34000/69632000 34190/70022840 RAID group /saggy/plex0/rg1 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.10 8 1 2 FC:A 34000/69632000 34190/70022840 data 8.11 8 1 3 FC:A 34000/69632000 34190/70022840 data 8.12 8 1 4 FC:A 34000/69632000 34190/70022840 data 8.13 8 1 5 FC:A 34000/69632000 34190/70022840 Spare disks RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- Spare disks for block or zoned checksum volumes spare 8.14 8 1 6 FC:A 34000/69632000 34190/70022840 The spare is not even good enough if one of the ZCS drives fail! Please let me how bad this is. Thanks! Steve **** ASI Solutions Disclaimer **** The material transmitted may contain confidential and/or privileged material and is intended only for the addressee. If you receive this in error, please notify the sender and destroy any copies of the material immediately. ASI will protect your Privacy according to the 10 Privacy Principles outlined under the new Privacy Act, Dec 2001. This email is also subject to copyright. Any use of or reliance upon this material by persons or entities other than the addressee is prohibited. E-mails may be interfered with, may contain computer viruses or other defects. Under no circumstances do we accept liability for any loss or damage which may result from your receipt of this message or any attachments. **** END OF MESSAGE ****

1 0

RE: mixing BCS and ZCS in the same volume
by Timothy Naple 14 Mar '03

14 Mar '03

Steve, You can add 36GB BCS (X231) drives to a ZCS volume, but you cannot use the 36GB BCS (X231) drives as spares for 36GB ZCS drives. The 36GB BCS disk size is smaller (34.0GB for BCS compared to 34.5GB for ZCS). The way you are running right now, you have no spares if one of your 36GB ZCS (X225A) drives fails. You should find another spare large enough to accommodate the 34.5GB requirement of your X225A's. I hope this helps, Tim Timothy Naple tnaple(a)berkcom.com 510-644-1599 VOICE 510-644-1598 FAX 888-812-9040 TOLL-FREE Berkeley Communications Corporation 2990 San Pablo Avenue Berkeley, CA 94702 http://www.berkcom.com -----Original Message----- From: steve(a)hq.newdream.net [mailto:steve@hq.newdream.net] Sent: Thursday, March 13, 2003 6:12 PM To: toasters(a)mathworks.com Subject: mixing BCS and ZCS in the same volume Is it bad to mix BCS and ZCS drives in the same volume? A person from a reputable Network Appliance supply company told me that this will probably most definitely lead to data corruption. Please read below, I know its a lot of information but it should be useful for someone to give me a straight answer. before we just had one FC9 shelf of seven 36GB ZCS disks on our F740 filer slot 8: Fibre Channel Host Adapter 8 (Qlogic 2100 rev. 3, 32-bit, L-port, <UP>) 0: SEAGATE ST136403FC NA11 34.5GB 512B/sect 1: SEAGATE ST136403FC NA11 34.5GB 512B/sect 2: SEAGATE ST136403FC NA11 34.5GB 512B/sect 3: SEAGATE ST136403FC NA11 34.5GB 512B/sect 4: SEAGATE ST136403FC NA11 34.5GB 512B/sect 5: SEAGATE ST136403FC NA11 34.5GB 512B/sect 6: SEAGATE ST136403FC NA11 34.5GB 512B/sect now we have another FC9 shelf of 36GB BCS disks but the sizes don't match slot 8: Fibre Channel Host Adapter 8 (Qlogic 2100 rev. 3, 32-bit, L-port, <UP>) 0: SEAGATE ST136403FC NA11 34.5GB 512B/sect 1: SEAGATE ST136403FC NA11 34.5GB 512B/sect 2: SEAGATE ST136403FC NA11 34.5GB 512B/sect 3: SEAGATE ST136403FC NA11 34.5GB 512B/sect 4: SEAGATE ST136403FC NA11 34.5GB 512B/sect 5: SEAGATE ST136403FC NA11 34.5GB 512B/sect 6: SEAGATE ST136403FC NA11 34.5GB 512B/sect 8: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 9: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 10: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 11: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 12: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 13: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect Before we added the second shelf I had it setup very nicely with a boot volume of 2 disks, data volume of 4 disks (w/one spare) and raid group size=7. Volume boot (online, normal) (zoned checksums) Plex /boot/plex0 (online, normal, active) RAID group /boot/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.0 8 0 0 FC:A 34500/70656000 34732/71132960 data 8.1 8 0 1 FC:A 34500/70656000 34732/71132960 Volume saggy (online, normal) (zoned checksums) Plex /saggy/plex0 (online, normal, active) RAID group /saggy/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.2 8 0 2 FC:A 34500/70656000 34732/71132960 data 8.3 8 0 3 FC:A 34500/70656000 34732/71132960 data 8.4 8 0 4 FC:A 34500/70656000 34732/71132960 data 8.5 8 0 5 FC:A 34500/70656000 34732/71132960 Then a person with very little experience (my boss) added another shelf to the data volume but without first checking the types of disks he was adding, etc. He added NETAPP X231 (BCS) disks to this ZCS voume. These are block checksum disks of a different physical specs. Now it looks like this. Volume saggy (online, normal) (zoned checksums) Plex /saggy/plex0 (online, normal, active) RAID group /saggy/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.2 8 0 2 FC:A 34500/70656000 34732/71132960 data 8.3 8 0 3 FC:A 34500/70656000 34732/71132960 data 8.4 8 0 4 FC:A 34500/70656000 34732/71132960 data 8.5 8 0 5 FC:A 34500/70656000 34732/71132960 data 8.6 8 0 6 FC:A 34500/70656000 34732/71132960 data 8.8 8 1 0 FC:A 34000/69632000 34190/70022840 data 8.9 8 1 1 FC:A 34000/69632000 34190/70022840 RAID group /saggy/plex0/rg1 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.10 8 1 2 FC:A 34000/69632000 34190/70022840 data 8.11 8 1 3 FC:A 34000/69632000 34190/70022840 data 8.12 8 1 4 FC:A 34000/69632000 34190/70022840 data 8.13 8 1 5 FC:A 34000/69632000 34190/70022840 Spare disks RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- Spare disks for block or zoned checksum volumes spare 8.14 8 1 6 FC:A 34000/69632000 34190/70022840 The spare is not even good enough if one of the ZCS drives fail! Please let me how bad this is. Thanks! Steve

1 0

RE: mixing BCS and ZCS in the same volume
by Strange, Steve 14 Mar '03

14 Mar '03

Steve, There is no risk of data corruption because of the mixed drive types. 520B/sector drives can safely be used in a zoned-checksum volume. The extra 8 bytes per sector simply go unused. You are correct that the spare you have is not large enough to be used as a replacement for any of the 34.5GB-usable drives. I suggest finding a 34.5GB 512B/sector drive to use as a spare, instead of the 34GB spare you currently have. Apologies that the drives are slightly different sizes. 520B/sector formatting was introduced around the time that "36" GB drives were most common, and when they were formatted for 520B/sector instead of 512, the resulting space wasn't quite enough to allow for 34.5GB of usable space. If we had it to do over again, we'd likely have made the 512B/sector drives slightly smaller, so all "36" GB drives would be the same usable size, and this issue wouldn't arise. We've moved away from ZCS for currently-shipping systems, although I know that doesn't help you on this particular system. Steve -----Original Message----- From: steve(a)hq.newdream.net [mailto:steve@hq.newdream.net] Sent: Thursday, March 13, 2003 6:12 PM To: toasters(a)mathworks.com Subject: mixing BCS and ZCS in the same volume Is it bad to mix BCS and ZCS drives in the same volume? A person from a reputable Network Appliance supply company told me that this will probably most definitely lead to data corruption. Please read below, I know its a lot of information but it should be useful for someone to give me a straight answer. before we just had one FC9 shelf of seven 36GB ZCS disks on our F740 filer slot 8: Fibre Channel Host Adapter 8 (Qlogic 2100 rev. 3, 32-bit, L-port, <UP>) 0: SEAGATE ST136403FC NA11 34.5GB 512B/sect 1: SEAGATE ST136403FC NA11 34.5GB 512B/sect 2: SEAGATE ST136403FC NA11 34.5GB 512B/sect 3: SEAGATE ST136403FC NA11 34.5GB 512B/sect 4: SEAGATE ST136403FC NA11 34.5GB 512B/sect 5: SEAGATE ST136403FC NA11 34.5GB 512B/sect 6: SEAGATE ST136403FC NA11 34.5GB 512B/sect now we have another FC9 shelf of 36GB BCS disks but the sizes don't match slot 8: Fibre Channel Host Adapter 8 (Qlogic 2100 rev. 3, 32-bit, L-port, <UP>) 0: SEAGATE ST136403FC NA11 34.5GB 512B/sect 1: SEAGATE ST136403FC NA11 34.5GB 512B/sect 2: SEAGATE ST136403FC NA11 34.5GB 512B/sect 3: SEAGATE ST136403FC NA11 34.5GB 512B/sect 4: SEAGATE ST136403FC NA11 34.5GB 512B/sect 5: SEAGATE ST136403FC NA11 34.5GB 512B/sect 6: SEAGATE ST136403FC NA11 34.5GB 512B/sect 8: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 9: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 10: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 11: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 12: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect 13: NETAPP X231_ST336704FC NA32 34.0GB 520B/sect Before we added the second shelf I had it setup very nicely with a boot volume of 2 disks, data volume of 4 disks (w/one spare) and raid group size=7. Volume boot (online, normal) (zoned checksums) Plex /boot/plex0 (online, normal, active) RAID group /boot/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.0 8 0 0 FC:A 34500/70656000 34732/71132960 data 8.1 8 0 1 FC:A 34500/70656000 34732/71132960 Volume saggy (online, normal) (zoned checksums) Plex /saggy/plex0 (online, normal, active) RAID group /saggy/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.2 8 0 2 FC:A 34500/70656000 34732/71132960 data 8.3 8 0 3 FC:A 34500/70656000 34732/71132960 data 8.4 8 0 4 FC:A 34500/70656000 34732/71132960 data 8.5 8 0 5 FC:A 34500/70656000 34732/71132960 Then a person with very little experience (my boss) added another shelf to the data volume but without first checking the types of disks he was adding, etc. He added NETAPP X231 (BCS) disks to this ZCS voume. These are block checksum disks of a different physical specs. Now it looks like this. Volume saggy (online, normal) (zoned checksums) Plex /saggy/plex0 (online, normal, active) RAID group /saggy/plex0/rg0 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.2 8 0 2 FC:A 34500/70656000 34732/71132960 data 8.3 8 0 3 FC:A 34500/70656000 34732/71132960 data 8.4 8 0 4 FC:A 34500/70656000 34732/71132960 data 8.5 8 0 5 FC:A 34500/70656000 34732/71132960 data 8.6 8 0 6 FC:A 34500/70656000 34732/71132960 data 8.8 8 1 0 FC:A 34000/69632000 34190/70022840 data 8.9 8 1 1 FC:A 34000/69632000 34190/70022840 RAID group /saggy/plex0/rg1 (normal) RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- parity 8.10 8 1 2 FC:A 34000/69632000 34190/70022840 data 8.11 8 1 3 FC:A 34000/69632000 34190/70022840 data 8.12 8 1 4 FC:A 34000/69632000 34190/70022840 data 8.13 8 1 5 FC:A 34000/69632000 34190/70022840 Spare disks RAID Disk Device HA SHELF BAY CHAN Used (MB/blks) Phys (MB/blks) --------- ------------ --------------- ---- -------------- -------------- Spare disks for block or zoned checksum volumes spare 8.14 8 1 6 FC:A 34000/69632000 34190/70022840 The spare is not even good enough if one of the ZCS drives fail! Please let me how bad this is. Thanks! Steve

1 0

← Newer
1
2
3
4
5
6
7
...
11
Older →

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 March 2003