toasters June 2014

toasters@lists.teaparty.net

26 participants
19 discussions

Fabrica metro cluster with ds14mk4 and ds2246
by Milazzo Giacomo 16 Jun '14

16 Jun '14

Hi everybody Can fc path coming fm atto bridges share the same iinitiators on the fas used also by the fc disk loop? Following an head swap fm 3160 to 8040 with 8.2.1 I've zoned a store zone with the fc port used by atto and the 4 initiators on 8040. Exactly as done with port coming from fc loops The fas starts to message a lot of timeouts and errors related to the sas disks then panic Of course the fc loops are on fc ports forced at 4 Gbps while the at to negotiate at 8. Can be this the.issue? Should I.have to.use a separate hba just for sas and another one for fc disk loops? Reading docs this don't seem a requisite. Regards Inviato da Samsung Mobile

3 6

snapmirror, flexclone and inconsistent volumes
by David Byte 13 Jun '14

13 Jun '14

Well, this is a fun one. I do some work for a shop from time to time that has a filer that is no longer under a maintenance contract. The other day, one of the guys saw condensation on the light fixtures and turned off the A/C for the room and left for the day. Needless to say, bad juju and a triple disk RAID failure.. So here we are today, I've managed to get the RAID back online (disk unfail, WAFL_check/wafliron) and have purchased some extra drives to replace the now non-existent spares pool. Here's the challenge. The primary data volumes were replicated off site on a nightly basis with snapmirror. There is a flexclone of one of the volumes where important data had been updated (don't ask) that had not been replicated. I have all the volumes online, but get errors on the console about inconsistencies when reading from all of them (there were severe media errors.) What I want to know... Is there some way to use snapmirror to bring the backing volumes back to a consistent state (including the snapshot backing the flexclone) and hopefully save their flexclone volume as well? Thoughts?'

1 0

Need Help - 100% ANY CPU
by Kalra, Manish 11 Jun '14

11 Jun '14

Hey Guys, I had an issue with one of my 7-mode filers (ONTAP 8.1.2) where all of the CPUs were 100% and there was no indication of any volume doing more IOs. It took down everything on the filer and DBs crashed. During the issue the output of wafltop show -v cpu: aggr_sas02_64:prod_volume1:nfsv3:WAFL_READDIR: consuming 300% CPU When we checked on the hosts mounting this volume we found out that there was one volume "deploy01" which was migrated to new c-mode filer but the mount was not fixed on those 3 hosts. So the NFS stack on those 3 servers was hung. The question now we have is: 1. Why would NFS stale on 3 hosts took down the filer? 2. Is there a BUG around this situation which someone has experienced before? Any help/clues would be greatly appreciated. Manish

1 0

Oracle Volumes and Compression
by Peta Thames 11 Jun '14

11 Jun '14

Hi, I've read that compression works well on volumes hosting Oracle DB data, but I haven't seen it in use anywhere in the wild yet. Is anyone using compression, and if so, what have been your experiences? Cheers, Peta

2 1

AW: FCP Ports operational down
by Alexander Griesser 09 Jun '14

09 Jun '14

I've done that several times so far and unfortunately it doesn't help... Any other ideas? If at least nothing obvious seems to be wrong, I think I'll open a support case on that issue. Alexander Griesser System-Administrator ANEXIA Internetdienstleistungs GmbH Telefon: +43-5-0556-320 Telefax: +43-5-0556-500 E-Mail: ag(a)anexia.at<mailto:ag@anexia.at> Web: http://www.anexia.at<http://www.anexia.at/> Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601 Von: S Oda [mailto:soda@socal.rr.com] Gesendet: Montag, 09. Juni 2014 18:05 An: Alexander Griesser Betreff: Re: FCP Ports operational down Try stopping and starting the FCP service. On 6/9/14, 6:35 AM, Alexander Griesser wrote: Hi there, I'm having a hard time currently connecting my new FAS8020 to our existing fabric - the FCP ports just won't get online and I'm not sure where exactly the problem is and how to debug that since I'm more the NAS guy not having much practical experience with FCP configurations, so here's the deal for all of you who are wanting to help. FAS8020 Clustered Data Ontap 8.2.1 Brocade DS4900B Brocade Switchport: > portcfgshow 0 Area Number: 0 Speed Level: AUTO(HW) Fill Word: 0(Idle-Idle) AL_PA Offset 13: OFF Trunk Port ON Long Distance OFF VC Link Init OFF Locked L_Port OFF Locked G_Port OFF Disabled E_Port OFF Locked E_Port OFF ISL R_RDY Mode OFF RSCN Suppressed OFF Persistent Disable OFF LOS TOV enable OFF NPIV capability ON QOS E_Port OFF Port Auto Disable: OFF Mirror Port OFF F_Port Buffers OFF Fault Delay: 0(R_A_TOV) NPIV PP Limit: 126 > portshow 0 portIndex: 0 portName: portHealth: OFFLINE Authentication: None portDisableReason: None portCFlags: 0x1 portFlags: 0x1 PRESENT U_PORT LocalSwcFlags: 0x0 portType: 10.0 POD Port: Port is licensed portState: 2 Offline Protocol: FC portPhys: 4 No_Light portScn: 0 port generation number: 0 state transition count: 0 portId: 010000 portIfId: 4302100d portWwn: 20:00:00:05:1e:5b:f9:a3 portWwn of device(s) connected: Distance: normal portSpeed: N4Gbps LE domain: 0 FC Fastwrite: OFF Interrupts: 2 Link_failure: 0 Frjt: 0 Unknown: 0 Loss_of_sync: 0 Fbsy: 0 Lli: 2 Loss_of_sig: 2 Proc_rqrd: 5 Protocol_err: 0 Timed_out: 0 Invalid_word: 0 Rx_flushed: 0 Invalid_crc: 0 Tx_unavail: 0 Delim_err: 0 Free_buffer: 0 Address_err: 0 Overrun: 0 Lr_in: 0 Suspended: 0 Lr_out: 0 Parity_err: 0 Ols_in: 0 2_parity_err: 0 Ols_out: 0 CMI_bus_err: 0 Port part of other ADs: No FCP Adater on NetApp: Node: jetfire Adapter: 1b Description: Fibre Channel Target Adapter 1b (QLogic 8324 (8362), rev. 2, 16G) Physical Protocol: fibre-channel Maximum Speed: 16 Status: link not connected Substatus: ADAPTER UP Host Port Address: 0 Firmware Revision: 6.10.0 Data Link Rate (Gbit): 0 Fabric Established: false Connection Established: loop Mediatype: ptp Configured Speed: 4 Configured State: up Switch Port: : AFAIK, the 16Gbit/s adapter should be backwards compatible to 4Gbit/s, our switch only has 4Gbit/s but it supports NPIV which seems to be a requirement for running Clustered Data Ontap via FC. So, I have no idea what information I should provide upfront for someone who's wanting to help me through get through that, but I'm more than happy to send more info when requested. Thanks, Alexander Griesser System-Administrator ANEXIA Internetdienstleistungs GmbH Telefon: +43-5-0556-320 Telefax: +43-5-0556-500 E-Mail: ag(a)anexia.at<mailto:ag@anexia.at> Web: http://www.anexia.at<http://www.anexia.at/> Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601 _______________________________________________ Toasters mailing list Toasters(a)teaparty.net<mailto:Toasters@teaparty.net> http://www.teaparty.net/mailman/listinfo/toasters

1 0

AW: FCP Ports operational down
by Alexander Griesser 09 Jun '14

09 Jun '14

FCP service has been created: ANX04-AUTOBOTS::> fcp show -instance Vserver Name: vFCP_jetfire01 Target Name: 20:08:00:a0:98:58:75:bb Administrative Status: up ANX04-AUTOBOTS::fcp> adapter show Connection Host Node Adapter Established Port Address ------------ ------- ----------- ------------ jetfire 0c loop 0 jetfire 0d loop 0 jetfire 1a loop 0 jetfire 1b loop 0 smokescreen 0c loop 0 smokescreen 0d loop 0 smokescreen 1a loop 0 smokescreen 1b loop 0 8 entries were displayed. ANX04-AUTOBOTS::fcp> show Status Vserver Target Name Admin ---------- ---------------------------- ------ vFCP_jetfire01 20:08:00:a0:98:58:75:bb up ANX04-AUTOBOTS::fcp> ucadmin show Current Current Pending Pending Node Adapter Mode Type Mode Type Status ------------ ------- ------- --------- ------- --------- ----------- jetfire 0c cna target - - offline jetfire 0d cna target - - offline jetfire 1a fc target - - online jetfire 1b fc target - - online smokescreen 0c cna target - - offline smokescreen 0d cna target - - offline smokescreen 1a fc target - - online smokescreen 1b fc target - - online 8 entries were displayed. Alexander Griesser System-Administrator ANEXIA Internetdienstleistungs GmbH Telefon: +43-5-0556-320 Telefax: +43-5-0556-500 E-Mail: ag(a)anexia.at<mailto:ag@anexia.at> Web: http://www.anexia.at<http://www.anexia.at/> Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601 Von: Krieger, Benjamin [mailto:Benjamin.Krieger@netapp.com] Gesendet: Montag, 09. Juni 2014 15:54 An: Adam Levin; Alexander Griesser Cc: Toasters Betreff: RE: FCP Ports operational down Hi, Make sure that you have created an FC service for the vserver owning those LIFs. ::> fcp show If you don’t see an entry, then fcp create. Required to assign a WWNN and bring LIFs online. --Ben From: toasters-bounces(a)teaparty.net<mailto:toasters-bounces@teaparty.net> [mailto:toasters-bounces@teaparty.net] On Behalf Of Adam Levin Sent: Monday, June 09, 2014 9:50 AM To: Alexander Griesser Cc: Toasters Subject: Re: FCP Ports operational down I'm not too familiar with Brocades -- we have Cisco -- but the first thing that caught my eye was trunk port being enabled. I don't think you want that, especially also for NPIV. You need an F port or an N port, not an E port. E ports are expansion ports, for ISL (inter-switch linking), and connect two switches together. F ports are fabric ports, and are used for point-to-point connections (hosts, storage systems, etc). N ports are node ports, similar to F ports, used for point-to-point or switched fabrics. For you, Fx or G port types should also work, and I think Brocades have U ports too, but I'm not too familiar. -Adam On Mon, Jun 9, 2014 at 9:35 AM, Alexander Griesser <ag(a)anexia.at<mailto:ag@anexia.at>> wrote: Hi there, I’m having a hard time currently connecting my new FAS8020 to our existing fabric – the FCP ports just won‘t get online and I’m not sure where exactly the problem is and how to debug that since I’m more the NAS guy not having much practical experience with FCP configurations, so here’s the deal for all of you who are wanting to help. FAS8020 Clustered Data Ontap 8.2.1 Brocade DS4900B Brocade Switchport: > portcfgshow 0 Area Number: 0 Speed Level: AUTO(HW) Fill Word: 0(Idle-Idle) AL_PA Offset 13: OFF Trunk Port ON Long Distance OFF VC Link Init OFF Locked L_Port OFF Locked G_Port OFF Disabled E_Port OFF Locked E_Port OFF ISL R_RDY Mode OFF RSCN Suppressed OFF Persistent Disable OFF LOS TOV enable OFF NPIV capability ON QOS E_Port OFF Port Auto Disable: OFF Mirror Port OFF F_Port Buffers OFF Fault Delay: 0(R_A_TOV) NPIV PP Limit: 126 > portshow 0 portIndex: 0 portName: portHealth: OFFLINE Authentication: None portDisableReason: None portCFlags: 0x1 portFlags: 0x1 PRESENT U_PORT LocalSwcFlags: 0x0 portType: 10.0 POD Port: Port is licensed portState: 2 Offline Protocol: FC portPhys: 4 No_Light portScn: 0 port generation number: 0 state transition count: 0 portId: 010000 portIfId: 4302100d portWwn: 20:00:00:05:1e:5b:f9:a3 portWwn of device(s) connected: Distance: normal portSpeed: N4Gbps LE domain: 0 FC Fastwrite: OFF Interrupts: 2 Link_failure: 0 Frjt: 0 Unknown: 0 Loss_of_sync: 0 Fbsy: 0 Lli: 2 Loss_of_sig: 2 Proc_rqrd: 5 Protocol_err: 0 Timed_out: 0 Invalid_word: 0 Rx_flushed: 0 Invalid_crc: 0 Tx_unavail: 0 Delim_err: 0 Free_buffer: 0 Address_err: 0 Overrun: 0 Lr_in: 0 Suspended: 0 Lr_out: 0 Parity_err: 0 Ols_in: 0 2_parity_err: 0 Ols_out: 0 CMI_bus_err: 0 Port part of other ADs: No FCP Adater on NetApp: Node: jetfire Adapter: 1b Description: Fibre Channel Target Adapter 1b (QLogic 8324 (8362), rev. 2, 16G) Physical Protocol: fibre-channel Maximum Speed: 16 Status: link not connected Substatus: ADAPTER UP Host Port Address: 0 Firmware Revision: 6.10.0 Data Link Rate (Gbit): 0 Fabric Established: false Connection Established: loop Mediatype: ptp Configured Speed: 4 Configured State: up Switch Port: : AFAIK, the 16Gbit/s adapter should be backwards compatible to 4Gbit/s, our switch only has 4Gbit/s but it supports NPIV which seems to be a requirement for running Clustered Data Ontap via FC. So, I have no idea what information I should provide upfront for someone who’s wanting to help me through get through that, but I’m more than happy to send more info when requested. Thanks, Alexander Griesser System-Administrator ANEXIA Internetdienstleistungs GmbH Telefon: +43-5-0556-320<tel:%2B43-5-0556-320> Telefax: +43-5-0556-500<tel:%2B43-5-0556-500> E-Mail: ag(a)anexia.at<mailto:ag@anexia.at> Web: http://www.anexia.at<http://www.anexia.at/> Anschrift Hauptsitz Klagenfurt: Feldkirchnerstraße 140, 9020 Klagenfurt Geschäftsführer: Alexander Windbichler Firmenbuch: FN 289918a | Gerichtsstand: Klagenfurt | UID-Nummer: AT U63216601 _______________________________________________ Toasters mailing list Toasters(a)teaparty.net<mailto:Toasters@teaparty.net> http://www.teaparty.net/mailman/listinfo/toasters

1 0

Reallocate in ONTAP 8.2
by Peta Thames 05 Jun '14

05 Jun '14

Hi, Reallocate has been like a magic performance-improving wand for me in the past (and thanks to members of this list for that). However I believe later versions of ONTAP have lessened the need for regular reallocation. Is that the case, or is it worth running regular reallocate measures? What's people's experience with ONTAP 8.2 and reallocation? Cheers, Peta

1 0

Backing up NetApp running CDOT with TSM
by jamesr 03 Jun '14

03 Jun '14

All, I've done a lot of browsing around of both NetApp resources and TSM resources and can't find an answer to this. This has lead me to this mailing list! We are currently deploying a NetApp FAS 8040 with around 500TB of disk behind it and we use TSM as a backup solution. Does anyone have a similar setup? How are you backing it up with TSM? Do you have configuration notes you'd be willing to share? We've looked at Snapdiff but this is unsupported with CDOT presently (http://www-01.ibm.com/support/docview.wss?uid=swg21633171) So I think we are looking down the NDMP route - however I'm not sure of exact implementation details of this... do we need to dedicate tape drives to this? What are its limitations? How do file restores work? etc. etc. etc. Anyone that can provide any assistance would be greatly appreciated. Cheers.

3 2

SAS controller limitations.
by Jeff Cleverley 01 Jun '14

01 Jun '14

Greetings, A while back there was a known issue with the SAS controller. Each chip that controlled 2 ports would only address 128 devices. That included shelf controllers so technically 5 ds2246 shelves would exceed the limit and could cause some slowness. I was told that they changed this behavior in 8.1 and possibly upgrading the firmware, so that it only kept active devices in the list. I was told those devices would be removed when not active and so there was no need to keep the smaller shelf limit. Is this the case, or do I still need to try to limit each chip to 5 shelves or less? I'm running 8.1.2P4 and the cards are running 01.11.00.00. Thanks, Jeff -- Jeff Cleverley Unix Systems Administrator 4380 Ziegler Road Fort Collins, Colorado 80525 970-288-4611

1 1

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 June 2014