Tag Info

New answers tagged

0

Battery and flash-backed write caches (BBWC/FBWC) are very important when working with HP Smart Array controllers. Performance tends to suffer without them. I suspect you may have a few things happening here... RAID5 is typically a no-no. Without any write cache, you're susceptible to I/O wait and disk rotational latencies during writes. Whatever is ...


0

Now I have tried with 32GB (2x8GB + 4x4GB) register memory, using only one CPU (E5520). According to specs the limit is 24GB (6x4GB). The system has been running very stable for 4 months now. I still really want to see if 32GB modules gets accepted. :-)


1

Lots of similar questions in a short period... Ubuntu isn't well-suited to vendor-specific hardware support. The HP ProLiant N54L server you have has binary drivers for the storage array that are only available for Windows and Red Hat. See: Cant find my harddrives in ubuntu installation? The RAID controller in this particular Microserver is a fakeraid ...


0

If the disks you're using are not HP drives and you're using a Gen8 ProLiant server, there's no guarantee that the disks will be recognized or will even work reliably. Your best option is to use actual HP branded/supported drives in the server. See: Third-party SSD solutions in ProLiant Gen8 servers


1

Silly question, but have you checked the firmware on your machine and its components? I ask because the changelogs for the DL320e Gen8 firmware include valuable nuggets like: Resolved a processor issue where certain instructions and conditions could result in an application-level error, blue screen, kernel panic, or other unpredictable system ...


0

You'll want to run BIOS upgrades before attempting this transition. Were you upgrading from an Intel 5500-series CPU? Did you previously have two CPUs? When the fan are running loudly, are there any messages on the screen? What do the server's health LEDs display on the front panel? Intel 5600-series CPUs are compatible with this platform with several ...


0

Anaconda asks DBus to check with NetworkManager if it's not connecting or if it's not connected already, thus, if there is network connectivity (so it won't ask about VNC). DBus will talk to NM but it seems that the call to _get_proxy (which creates a proxy object to talk to DBus) failed. So it seems there is a problem with DBus, but it's not clear just ...


7

Just don't use the Sun rack!! It really can't be converted. Starting a new project with a round-hole non-standard rack today is irresponsible. Racks are not expensive new. Racks can be purchased used or refurbished, if cost is a concern. If you're going through the steps to rack mount equipment, you may want to future-proof it a bit. A more ...


6

The EIA-310 standard does have provision for round holed racks, however the Server industry has pretty much stopped supporting them. Square holed racks are far easier to build for and are much more tolerant of slight variances in rail build. Thus, for vendors who have to sell into a market where buyers can have racks from any of a hundred suppliers it makes ...


0

Is there any chance you can try this with a soft keyboard on the Mac instead? I've always had problems with remote Ubuntu installations over IPMI and lower-end ILO solutions.I usually have to resort to an on-screen keyboard on the ILO/IPMI side or on the local OS side. (sorry, I can only give these instructions in English...) Open the Language & Text ...


1

On an HP printer, a PCL driver (either 5 or 6) will likely work the best. The intrepreters on a HP device handle PCL much better than PS [they developed the language, so they better!]. If you are just doing regular windows printing then PCL6 would be my suggestion. The PS interpreters on HP devices are not great and I would generally avoid it. Other types ...


3

Ready for Rebuild is a bad status if you're using a parity RAID level, like 5 or 6. It means that you likely have read errors on another drive in the array... e.g. another failing drive. If the system is still online your best option is to recover data or rebuild. There's no good fix for this, and definitely not much you can do to debug. See the following: ...


4

With DAC cables, I usually let the switch dictate compatibility. Go with what is known-good on the switch and work from there. The other part of this is your balking at prices. Part of working in this field is vendor management and establishing vendor relationships. There's almost always a better pricing option; especially with HP equipment. Try some of ...


4

No, there is no equivalent on Procurve gear. If you baulk at HP prices, your best option is probably to look for a third-party supplier that will sell compatible parts with the SFP+'s coded as HP.


23

Wut?!? This is crazy, you know? It's unsupported and likely not good for your hardware. You should be using equally-spec'd processors in your server. I suppose you couple simply execute your application in a cgroup or shield (or via taskset) that only contains the CPU cores you wish to use. You can also pin memory access to that CPU with numactl. But ...


2

Raid1 is just a mirrored disk, so if you enable raid1 and have your 1st disk as primary and the 2nd as the mirror, no you wont loose any data. Did this yesterday on a dl320 and it was fine


2

You can't remove configuration rules that set values with the [no] syntax; the way to remove them is to set them back to their default value, which may involve RTFM. So if you have the two rules DarkMoon gives: aaa port-access mac-based 9 aaa port-access mac-based 9 addr-limit 32 they are disabled with: no aaa port-access mac-based 9 aaa port-access ...


1

It's already off. Use show port-access mac-based to show the status of each port. What you have in that config is the address limit for each port, but have not enabled mac-based auth on any of them. Usually, for a port that has it enabled, you should see: aaa port-access mac-based 9 aaa port-access mac-based 9 addr-limit 32 The two lines do different ...


0

Yes, this switch can act as an NTP server. There are different ways to set it up, namely client/server, symmetric peers, multicast and broadcast modes. Most often you will likely set it up in a client/server mode. For this to work you need to set up NTP synchronization on the switch itself and then it can act as a server for other devices on your LAN. ...


1

The defaults will do what you need without issue... That is, there's nothing in the default configuration that would preclude proper operation of vSphere 5.5. You may want to modify a few other BIOS settings, depending on your workload and usage patterns: Power Management: My applications care about latency, so I tend to disable the power and CPU-scaling ...


2

Sure, this is an easy process... Once the rebuild is done, you can either use the "Unused Space" to create another Logical Drive (e.g. Logical Drive 3), or you can Extend Logical Drive 2 to utilize the extra space. Once this is done, you'll see a larger block device available to your operating system. If this is Windows, the free space will be ...


1

I have BL460c G1, and I simply use standard settings, but ensure Intel VT technologies, and Execute Disable Bit, or INTEL TXT (if you have) are enabled, they are madatory for ESXi 5.5


1

The "Error: Line #1: syntax error near "?>"" message is probably due to the .xml file being written encoded as Unicode and it needs to be ASCII/ANSI. Open the file in Notepad and do a "save as". See what encoding it is saving it as.


2

256k+ - SQL Server, at least enterprise, tries to do IO in this size and allocate clusters (64kb, 8 pages of 8kb) in groups of 4 to optimize IO. So, cluster size should be multiple of 256kb.


0

The issue is getting the correct drivers. Typically you will need the 'Full Feature' driver to use the fax functionality, often they will not release a server version of this but rather a Universal PCL5/6 driver. Even when you manage to get the full driver to install, the server will not be able to serve the fax to users any more than a PC can. This is what ...



Top 50 recent answers are included