Hi All,
Here is the situation:
- MD3860F with dual-controller
- Server has 01 single-port FC HBA. So server has only 01 direct connection to MD3860F controller (no FC switch there).
Now we have 'Virtual Disk Not On Preferred Path" alert. Is there any disruption if we manually change the Ownership/Preferred Path of the virtual disk (disk group)?
Thank you,
Hello.
We have tl2000 with IBM ULT3580-TD4 LTO4 drive, FC interface, Finistar 4 Gbit GBIC, connected to HP DL180 G6 server with P410 RAID, 7 WD Red drives, RAID5, HP LPE1150VM FC HBA, Win 2008 R2.Tape drive firmware C7QH.
With BE 2012, backup speed is about 100-110 mbyte/sec, 6200-6500 mbytes / minute. It's quite less that it should be for LTO4. Please advise, is it possible to increase the backup speed?
| C7QH |
Hi,
We have a MDI3000 that is connected to a power edge server with 4 nics.
Currently the SAN is connected with only one ISCSI connection (RAID controler module 0 port 0), becuase we have some read speed issue and the host server has a free nic, we think it would be a good idea to connect the second ISCSI from module 0 to the free nic of power edge server.
For doing so does these steps enough?
1. Give the RAID contorler Module 0, port 1 IP: 192.168.131.101
2. Give the free nic of the power edge server IP: 192.168.131.100
3. Connect the free ISCSI port of MDI3000 to the nic in the step 2.
4. During the connecting of the second ISCSI port to the host server no data intreaption will occur.
are these steps enough/correct?
our current configurartion:
After:
I want to get the most utilization for backups out of my PowerVault. I have an old MD1000 that seems to burn up hard drives. For this reason, I will maintain 3 hot spares in the array. With the remaining 12 disks, is it better to have a single 12-disk RAID6, or two 6-disk RAID6? I/O will be bursty, but as this is a backup server, performance is not critical.
Additionally, how much of the virtual disk volume can I safely consume without creating an issue? If I have a 5TB virtual disk, mount it to a Windows server and create a 5TB volume, can I place a 5TB fixed VHDX on it without issue? Or do I need to leave 10% free space?
Hi there,
I have a very serious issue with an MD3800i (dual controllers) that I have just purchased. I have an MD1200 expansion array attached. I am getting expected write performance (800mb/sec+) however I'm getting around 2mb/sec reads.
I have tried changing cache settings, creating smaller LUNs, multiple host OS's (xen/vmware/hyper-v), I've tried MPIO and fixed/single target and nothing makes a difference. I have dual controllers and I have tried fixing to each controller individually and still no change (so isn't a controller at fault). I have changed the block size, I have taken the switch out of it and connected directly to the SAN, I have tried disabling prefetch, everything I can think of. This is stopping me from configuring my new environment and is making my investment seem like a bad choice.
I have noticed that when I read a file from the SAN, the disk latency flies up to around 30ms. I have also noticed that for around 1s at the very start, the read speed is very quick and then drops down to the awful performance.
I have tried to do some fixes I found on the internet regarding setting the cacheFlushStart and cacheFlushStop however the cacheFlushStop command has been deprecated in a later version of the firmware so I cannot try this fix. The fix by wallewek sounded promising until the firmware wouldn't let me do it!
I cannot believe that this SAN can only read at 2mb/sec (sometimes I can get as high as 20-30mb/sec with MPIO, but this is still not acceptable!). I have a very old iSCSI SAN with gigabit links that max out at 100mb/sec+ so something is wrong here.
Any help would be appreciated - I'm lost as to what to do now next now as I feel like I've tried everything possible. I have spent over a week trying everything I can see and think of and I'm getting nowhere.
Thanks
Does anyone have input on connecting a perfectly good and serviceable 4/1G Md3220i to a 10G network with ESXi? My initial though it to create 4 Vmkernel ports on the same 10G nic, and then assign each one a separate IP address.
Anyone have thoughts on that?
Have an odd question, is it possible to use and MD3400 as an expansion for a MD3200i?
Have had a long running issue with the MD3400 and linux/xenserver. We are in need of expanding storage, and had a thought if we could do this, I would rid myself of a problem that Dell, nor anyone else, has been able to resolve.
If I can, will I have to rebuild all the storage on the MD3400, or will the configs import into the 3200i?
I can see where this could work, but also where it cause problems... Thx.
How do I fail it so I can put in the new one - or do I just wait for it to fail? Thank you for your help.
What are the steps to adding the new MD3800i to my current network? This unit will be replacing my MD3000i. Also, do I uninstall the current MDSM that is loaded and then load the new version that came with the MD3800i? I want to add the new unit into the production WITHOUT bring down my current MD3000i. The plan is to move the VM's off the 3000 to the new 3800.
Thanks.
I have a PowerVault MD3000i that's been showing signs of aging, I guess. In the last 3 years I had to replace 7 HDDs and 2 controller module's batteries, the second battery instance being last night. After I was done I re-checked the health of the array and it was optimal, the flashing amber light changed to blue and I went home. This morning the same battery on the same controller module is in error, so I am guessing that the module itself is burning the battery or the main board is the culprit.
I'm hoping that I can gather some data and send it to a Dell tech so they can tell me what they see.
Also - before I started the procedure I "stopped the I/O" by turning off the network switch that connects the SAN's interfaces with the servers (2 PowerEdge 2950s). This morning all virtual machines (HyperV) except one were off, but the hosts were online. Do the VMs turn off when they don't find the path to the SAN? I turned them on and they all came back with the "windows didn't shut down normally" message. I can't find anything explicit on the Event Viewer.
Thanks.
Hi
Are there MIB files for the DR's and is it possible to query them via SNMP?
/Peter
Hi all,
I have a PowerEdge T420 running Server2012 R2 and am trying to connect my Powervault 114X with a RD1000 Hard Disk drive that I bought in 2010.
The server reports a USB problem and will not connect. Device Manager reports "Windows has stopped this device because it has reported problems (Code 43)"
I can't find any driver or firmware updates for this unit.
I have restarted the server various times with the 114X powered up, but always the same error.
Any ideas what is the problem might be?
Thanks
Hi fellow storage lovers,
I have two quick questions.
1) Could you point me in the right direction to documentation on the correct procedure to access the sliding disk bays on this unit WHILE it is running? (if that is possible?)
2) Would running a copy of MD Storage Array Software on the network attached to this device for the sole purpose of getting hard drive information (number of drives, type, size, etc) effect the already running copy?
Thanks in advance!!
What model number of LTO5 or LTO6 drives is compatible with this 124T array? My drive is bad (showing ERROR: driveTAflag(31) hardware B). It came with an LTO6 drive. passes all other tests.
Ive try dell support but i just got it off of ebay (brand new) and they say they need dell invoice which i don't have. Now stuck with buying a refurbished drive to minimize loss.
So which drives are compatible with this that i can use and where can i get them. thanks! i really need to get this thing working
I'm trying to ship a ML6010 but I'm missing the device that holds the robot arm steady during shipping. Does anyone know where I can get one or what else I need to do to ship a ML6010?
I have rarely used the robotic library for back ups and now when I do, it's not working. When I perform a move of tape from drive to I/O I receive the message "HE: drive unload timeout - Code: F5". The LCD screen of the unit displays "Warning: Incomp. Medium".
I've seen an article how to manually release a magazine--that is simple enough but have not come to a resource that would help me in my case.
It is a LTO Ultrium 5-H tape drive (P/N 46X2478) in the sled. I have reseated the sled with no results.
I've searched and searched some more without any help from online resources.
Thanks
I am setting up a new MD3820i and while applying changes in the Configuration Utility I get the following error: Unknown Error (PVCF_EXCEPTION). The unit reports itself as optimal. What causes this and how can I move past it?
Thanks in advance for any assistance.
Building a Poweredge R810, SAS HBA card/cables and Powervault TL4000 tape library using Windows server 2012 R2. Device mgr shows IBM Total Storage 3573 Tape Library, location 2 (Bus Number 0, Target ld 3, LUN 1) but status of...
"Windows cannot initialize the device driver for this hardware (Code 37), Object Name not found."
SAS HBA and tape library were working fine on a PE 710.
Downloaded "Dell TL2000/4000, v.9.30, A11" file name DELL_TL2000-4000_A11_R266766.exe and installed but it says the "best driver software for your device is already installed".
System event viewer shows Event ID 219 - The driver \Driver\IBMCgFt2k12 failed to load for the device SCSI\Changer&Ven_IBM&Prod_3573-TL\7&132630dc&0&000301."
What else can I try to get the tape library to be seen correctly by the server?
Hi,
Currently our MD3000 is connected with only one ISCSI connection (RAID controler module 0 port 0), becuase we have some read speed issue and the host server has one more free nic, we think it would be a good idea to connect the second ISCS portI from module 0 to the free nic of power edge server.
Question
Configuring the IP for ISCSI modul 0 port 1 and connecting it to the free NIC of the same host as Module 0 port 0 is enough? or after connection has been made we must configure the ISCSI initiator on the server 2008 R2 as well?
Thanks