Failover of fiber channel time

We have a v890 with 2 single channel hba's, Solaris 8, Veritas 4.1 (DMP), connected to a Clariion CX700 (Active/passive).
We are still in test mode for this configuration.
When we disconnect one of the fiber's at the HBA to simulate a failure, it takes almost 2 minutes before failover occurs and the other path is used.
Is this the expected timeout for failover to occur? I suspect this is controlled by DMP? Is there a way to make this happen quicker? Eventually this will be a db server and I'm concerned about the amount of time it takes to failover.

unlpugged fiber....
Apr 6 13:00:19 dbserv02 qlc: [ID 630585 kern.info] NOTICE: Qlogic qlc(1): Link OFFLINE
Apr 6 13:00:19 dbserv02 scsi: [ID 243001 kern.warning] WARNING: /pci@9,600000/SUNW,qlc@1/fp@0,0/ssd@w50060162306007c7,5 (ssd14):
Apr 6 13:00:19 dbserv02 SCSI transport failed: reason 'tran_err': retrying command
Apr 6 13:00:54 dbserv02 scsi: [ID 243001 kern.warning] WARNING: /pci@9,600000/SUNW,qlc@1/fp@0,0/ssd@w50060162306007c7,5 (ssd14):
Apr 6 13:00:54 dbserv02 SCSI transport failed: reason 'timeout': retrying command
Apr 6 13:01:49 dbserv02 fctl: [ID 517869 kern.warning] WARNING: 186=>fp(1)::OFFLINE timeout
Apr 6 13:01:52 dbserv02 vxdmp: [ID 917986 kern.notice] NOTICE: VxVM vxdmp V-5-0-112 disabled path 118/0x70 belonging to the dmpnode 239/0x38
Apr 6 13:02:08 dbserv02 scsi: [ID 243001 kern.info] /pci@9,600000/SUNW,qlc@1/fp@0,0 (fcp1):
Apr 6 13:02:08 dbserv02 offlining lun=5 (trace=0), target=611113 (trace=2800004)
repeating stuff about offlining of LUNs......
just plugged in fiber....
Apr 6 13:03:43 dbserv02 qlc: [ID 630585 kern.info] NOTICE: Qlogic qlc(1): Link ONLINE

Similar Messages

What are the host network requirements for a 2012 R2 failover cluster using fiber channel?

I've seen comments on here regarding how the heartbeat signal isn't really required anymore - is that true? We started using Hyper-V in its infancy and have upgraded gleefully every step of the way. With 2012 R2, we also upgraded from 1gb iSCSI
to 8GB Fiber Channel. Currently, I have three NICs in use on each host. One for "No cluster communication" on it's own VLAN. Another for "Allow cluster network communication on this network" but NOT allowing clients, on
a different VLAN. And lastly the public network which allows cluster comms and clients on it (public VLAN).
Is it still necessary to have all three of these NICs in use? If the heartbeat isn't necessary any more, is there any reason to not have two public IPs and do away with the rest of the network? (two for fault tolerance) Does Live Migration
still use Ethernet if FC is available? I wasn't sure what all has changed with these requirements since Hyper-V first came out.
If it matters, we have 5 servers w/160GB RAM, 8 NICs, dual HBAs connected to redundant FC switches, going to two SANs. We're running around 30 VMs right now.
Can someone share their knowledge with me regarding the proper setup for my environment? Many Thanks!

Hi,
You can setup cluster with a single network but that leaves you with single point of failure on the Networking front, it is still recommended to have a heartbeat network.
Live migration would still happen though Ethernet, it has nothing to do with FC. Don't get confused, you had iSCSI for storage which used one of your VLAN and now you have FC for your storage.
Your hardware specs looks good. You can set up the following networks -
1. Public Network - Team two or more NICs (based on bandwidth aggregation)
2. Heartbeat Network - Don't use teamed Adaptor
3. Live Migration - Team two or more NICs (based on bandwidth aggregation)
Plan properly and draw guidelines to visualize and to remove single point of failure at all points.
Feel free to ask if you have some more queries.
Regards
Prabhash

Fiber Channel Error

Having a problem with the Fiber Channel on our Xserve RAID. The Admin Utility says that Fiber Link 1 is Down (and it's red), but everything is properly connected, I have restarted several times, unplugged and plugged back in, done a few upgrades to the firmware and it still says it's down.
Any ideas?
Thanks.

I swapped the Fibre cables, controller and reseated the card in the mac. Still nothing. I noticed under the Hardware Overview that there is not even an option for Fibre Channel. Is this due to the system running 10.3.9?
Thanks.

External Fiber Channel Raid options for New Xserve besides Promise V-Trak

I was wondering if anyone out there knows if there is an alternate option for a External Raid that can be hooked up via 4gb Fiber Channel to my new Xserve. I am really looking for a more cost efficient one than the 12K Promise set up from Apple. My Technology vendor was telling me that you have to use the Promise Raid as it is the only one that a Xserve will control due to Apple's licensing of its drivers etc. for its Fiber channel. Something tells me that this is not true.
Anyone have any recomendations for a Fiber channel Raid solution in the 6K dollar range?
thanks,
Dan

Find another vendor. Anyone that tells you the Promise is the only RAID that will work clearly doesn't understand the market.
As for recommendations, you need to include some idea of capacity, either usable or raw. I can find you a lot of options for $6K if you want a 100GB of data. Not so many if you want 100TB.

XSERV RAID FIBER CHANNEl

We have XSERV RAID MODEL A1009 with 4 Hard disk RAID 5.
Now I have installed CISCO 1000Base SX Module in 3 COM Switch & other same Module in XSERVER RAID. Now XSERVE RAID & switcted with Transeiver cable. Now I check RAID ADMIN TOOL FIber Channel Option the links shows down.
Let me know what is solution. Please also let me know how can I know that fiber link is up if any way.

Finisar makes a nice SFP to LC transceiver (available on the Apple Store for $99.95 for 2 Gbit and $119.95 for 4 Gbit) that you can use in the Xserve to allow it to accept LC optical cables. You'll need 2 of these, and then you can hook up LC optical cables.
I'll echo the other post -- be aware those ports are for fibre channel communication ONLY. You must hook it to a fibre channel card (HBA) in a host, or to a fibre switch (i.e. Cisco MDX 9000 series). You cannot hook up a network connection to those ports -- Xserve RAID provides data ONLY over fibre channel, it is not a NAS device. You must connect it to a host which can re-share the data over Ethernet, for example Xserve RAID with OS X Server which can reshare using AFP, SMB/CIFS, NFS, FTP, DAV, and more.

Kernel panic when connecting fiber channel

I recently inherited an Xserve RAID. I have an LSI fiberchannel card in a G4 running 10.5.5.
Anyway, I updated the firmware to version 1.5 and now everytime I connect the fiber channel cable I get a Kernel panic instantly and have to force power down.
I've tried 3 different cables on and every port on the mac and the xserve and get the same result.
any idea what's going on?

Well, i just upgrade to 10.8 and its solved the question.

IP over fiber channel

Idle curiosity question:
Does anyone know whether or not Netware or OES supports IP over fiber
channel?

If you dont use 802.1q (vlan):
You must create another fcip interface and use existing fcip profile, but when configuring peer-info
address you should use another ip address.
fcip profile 1
ip addres 1.1.1.1
exit
int fcip 1
use profile 1
peer-info ipaddress 1.1.1.2
int fcip 1
use profile 1
peer-info ipaddress 1.1.1.3

Use of Fiber Channel in Boot Camp with Windows XP

Hi All,
I am looking at getting a Mac Pro for a project that requires I use Windows XP under boot camp and wanted to know if there were any problems accessing a RAID array via the standard 4GB Fiber Channel adapter available through apple. I don't need to boot off this drive, just access data on it. Not a lot of information on the official apple support pages...
Thanks!

Yes, XP will be able to detect the expansion card if you download the appropriate drivers online. When bootcamp was first starting... this article circulated http://docs.info.apple.com/article.html?artnum=305401... summary, you should remove the fiber channel card before bootcamping. Since then the software has gone through the proper modifications that you can just plug the card in, boot into windows, and download the LSi driver for XP.
So you'll have no problem setting this up.
Cheers
Ron

OVM 3.1.1 Add new Fiber Channel LUN on the go

Hello everyone,
I've a question regarding Fiber Channel LUNs. I need to present a new Fiber Channel LUN to my 2 production servers and I would ask what's the best procedure to do this.
Usually I turn off servers present FCP LUN then turn servers on, but what if I can't poweroff servers?? There's a way to present FCP LUNs without powering off servers??
Thanks in advance. ;)

I have already met this problem.
In the Manager, you can right-click a VM Server and select "re-scan physical disks". That sounds like it would work, right?
Well, it has NEVER worked for me. Maybe someone can tell me what I'm doing wrong.
But what DOES work is:
Present the LUN to all servers.
Go to one of the servers, log in as root, and run: rescan-scsi-bus.sh
This makes the new LUN known to the server.
Then go back to the Manger, and do the "rescan physical disks" thing. This gime the new LUN will appear, and it even appears to ALL servers in the cluster. You're done. (It seems like sometimes the LUN will appear in the GUI after running rescan-scsi-bus.sh, without doing anything. But I'm not sure.)

Mac Pro w/ Fiber Channel, Can it sleep with XRaid attached?

I wanted to know if the new mac Pro can fall asleep when it has an xraid attached via its fiber channel card.
My Dual 2ghz G5 does not, this is why I'm asking.
Your comments are welcomed.

Not sure, but on the 'old' one,by design, NEITHER the Fiber card and Raid card will allow 'doze' but not 'deep sleep' of the Mac Pro in order to sustain connectivity, battery charge and redundancy, respectively.
I doubt that changed.
Anyone?
d

4Gb Xserve Raid Card to 2Gb Xraid Fiber Channel

Greetings All,
Does anyone know if a 2x4Gb Fiber channel card will auto-negotiate down to the 2Gb level of my Xraid?
Thanks,
-dave

Yes it does. I've done it and the newest Xserve is plug and play with the old Xserve RAID.
=Tod

Is my fiber channel card failing?

Is there a way to test a fiber channel card.
I believe mine is failing, but don't know how to confirm.
Every few days the fiber channel links to my Xserve Raid go down.
Raid Admin shows no trouble.
Usually it comes back after rebooting the Xserve.
This is making things extremely difficult, please help!

Here's my situation. At home we use 3 different Macs
all on the local wireless Airport network. One is a
G5 that I use in my home office, another is my wife's
G4 laptop and the other is a relatively new iMac - it
is the iMac that is a problem.
Is this a new Intel iMac?
Are you using WPA?
Some Intel iMacs have issues connecting to a WPA encrypted network which results in a slow network connection.
I have read that there is an issue with the Intel iMac and wireless speeds when using WPA?
Are you using WPA?
What happens if you change to WPA2 ?
Or WEP?
iFelix

Fiber Channel Storage Directly Connected to Nexus 5000

Can you connect Fiber Channel SAN storage directly to a Nexus 5000.
I swear I remember that you could not (at least when it first came out) and you had to connect it to a FC SAN switch. The docs look like it can act as a full fledge SAN switch, but it does say "Each Fibre Channel port can be used as a downlink (connected to a server) or as an uplink (to the data center SAN fabric)."
So It sounds like a server can connect to it directly via Fiber Channel, but you still need to connect it to a switch (i.e. data center SAN fabric).
Thanks for any help in clearing up this confusion.
Joe

You can connect any storage array and get it to work from a protocol perspective. But most storage OSMs have a certification process and you might want to check to see if the storage array you have is tested and certified for direct connect target.

Win 2012 R2 virtual fiber channel

Hello,
I enabled virtual fiber channel adapters on the physical Win 2012 R2 node.
Now I would like to enable virtual fiber channel on the virtual machine. I shut downed VM and added two virtual fiber channel adapters. I turned on VM and fcinfo gives me information about them.
Need to mention that I'm using default Windows WWNs.
Now I would like to connect some LUNs from NetApp to VM. I've added WWNs of VM to NetApp but LUN doesn't appear in the VM.
My setup:
Cisco UCS 5108 chassis
FI6248 fabric interconnects
B200M3 blades
I don't have Nexus switches.
flogi shows connected VM fiber channel to the fabric interconnect.
INTERFACE VSAN FCID PORT NAME NODE NAME
vfc809 100 0x290201 c0:03:ff:0b:b5:9f:00:0a c0:03:ff:00:00:ff:ff:00
My question is how to setup fc zoning on the fabric interconnect that VMs could connect to the NetApp storage via FC.
Is there any guide how to setup FC zoning by Cisco best practices?
Thank you

Hi Aleksandras
Maybe the situation / problem is the following:
1) you have e.g. 2 FC initators which are Hyper-V host based (eg. vhba0 and vhba1).
2) in addition you have the FC initators of the virtual hba's of the VM's.
You have a direct attached storage, and zoning is done on the FI. However, this automatic zoning only happens with FC initiators defined in the vhba's of the service profile, 1) above
UCS is totally unaware of the OS, and therefore of virtual hba's of the VM's.
That's why you see the flogi of the VM virtual hba's but no corresponding zone.
Solution: connect your UCS via a FC switch to the storage subsystem, and the zoning can be done for the Hyper-V host, as well as the VM's.

Can fiber channel be used for connect server to a storage in two nodes RAC

I decided to have an Oracle RAC cluster with only two nodes.
The storage I have is a sun 3510FC. Also have five fiber chanal ports. But my each server (Sunfire v240) has got one Ultra160SCSI multimode (SE/LVD) not any fiber chanal ports. I found that there is some FC card available to use but none fiber switch available. I want to install this FC cards to the servers and make a straight connection with fibre cable to the storage. Can this be asolution or do i have to have a fibre switch?

Hi
Irrespective of whether such a configuration is supported or not by Oracle.
Has anybody got 10gR2 RAC working on
a. 2 servers
b. shared storage with HBA and fibre channel cables
c. interconnect on LAN
(yes we know it is not supported by Oracle...but again we are not going production on that....just for demo/training purposes)
No FC switch
No ethernet switch
If anybody has got this working would be great to hear of 'issues to guard against'
Thanks

Failover of fiber channel time

Similar Messages

Maybe you are looking for