Friday, March 27, 2009

My HPTF session proposal has just been approved

The e-mail came in late this afternoon: my HPTF 2009 abstract has been approved. If all goes well, I should be there this year to present Monitoring and profiling Virtual Connect performance with HP-UX or Windows.

This session will show how to use cacti to characterize and monitor the performance of your virtual connects, and iperf to actually test the potential bandwidth. These are FOSS multiplatform tools, and they work well. I'll also add some few caveats that I stumbled upon when we deployed our VC infrastructure last year.

See you there.

Friday, March 20, 2009

Configuring SysFaultMgmt and RSP on HP-UX

I've stumbled upon the following howto:

I haven't gone through it entirely yet but it seems quite complete.

Good luck

Saturday, March 14, 2009

The ridiculous price of HP Enterprise Software

A colleague of mine told me recently that "Enterprise" software that's usually despised by IT technical staff often finds its way in corporations using steaks and hookers. To this I would add good golf appartus. Unfortunately for these software sellers, I'm not fond of steak, I'm married, and I don't play golf... So my software purchases are based on technical merit. And I must say that increasingly, Open Source software wins in this arena. But that's another story.

Some Enterprise software are actually useful to IT techies, and even if they're costly, they can actually make you save in the long run. One example I can think of is OpenView Performance Manager. I've been using it for years and it helped us plan equipment purchases and consolidate on VMs. Its price could be considered expensive by some, but personally I find it reasonable when I see how useful it's been for our environment.

I might have been naive, but I expected a cost similar to OVPM when I asked about the price of a storage management software from HP which I won't name here. As a SAN admin, that software would have been very useful to me. It would have helped me provision and plan for years ahead our disk space trends.

But my inquiry came back with a price tag that was the absolute highest, by a large degree, that I've ever encounted up until now in my 10+ years in IT.

Man, what are these guys smoking?

Do they expect someone to actually pay this price, especially in these dire economic times? Give me a break. The cost is so insane that no matter what that software does, unless it actually prints money, there's no way a lone IT admin such as me could even think of recommending this to management. My chances would be better endorsing the purchase of a few Ferraris, at least they'd be fun to drive.

Joel Spolsky once wrote an essay detailing how to price software. If I remember correctly, it boiled do to this: you set the pricing strategy depending on the what your target customer is ready to pay.

Did you know that HP recently asked their employees to lower their salary ? No wonder. I don't know who can purchase software at these prices in 2009. The target customer for the management software I asked for is either someone who's IT budget is in tens (if not hundreds) of millions and don't give a damn about the price of their software, or they're the kind who has a soft spot for Angus beef, exotic dancers and nice golf clubs. I'm none of that. Looks like I'll just have to spend a few weeks gluing together some FOSS software, along with a few homemade scripts, and I'll have my own management solution for free.

Thursday, March 12, 2009

Time for a small overhaul

As there are some code snippets here, I decided to change the layout of my blog to make the articles wider and on a white background, it should be easier to read. As for the origin of Technocrat-UX... yes, it used to be a pun on which, unfortunately, shut down last december.

The purpose in life of vxfsd

Currently, the #1 Google keyword that has readers stumbling on this blog is "vxfsd" and they spend some time checking out the entry where I tell how to reduce vxfsd usage on 11iv2. It figures; vxfsd often shows up first in top(1) and there is not much documentation on what it does, let alone a man page. I think an explanation of what this this daemon is used for is in order.

Under Unix, an inode is a structure that contains file metadata (size, position on disk, owner, etc). You can find a .pdf with an example inode here. HP-UX puts these inode structures in a cache to accelerate I/O performance. That area is called the inode cache. It is a dynamic cache, meaning that it grows or shrinks its size RAM depending on the needs.

Where does vxfsd come in? It's a daemon whose job is to scan the inode cache, freeing inodes which have not been referenced since a while. The more inodes there are to scan and free, the more it spins and consumes CPU time.

When there is intensive read I/O on a server, such as during a backup, all files are read, so a great bunch of inodes get cached in the progress, with the cache size growing quickly. It is normal under these circumstances to see vxfsd spinning once that I/O pressure is over; it's doing its job, scanning the cache and freeing the inodes that haven't been recently needed. It should slow down once it's done.

As I described in my earlier post, the kernel can be tuned to use a static inode cache, and this will make vxfsd stop needing to scan the cache. It can actually be a good idea to do this on low-end systems (especially VMs) that have few or slow CPUs, as vxfsd can consume a fair percentage of CPU time on these types of systems. The trade off is that you could have lower I/O performance since the cache will no longer be able to grow to meet a sudden I/O demand. I normally don't care about vxfsd on systems that have multiple processors, and let it live with its default settings.

Take care