Older blog entries for huge (starting at number 14)

Uh.

I just remembered my password.

Yes, 3 years after my last post ...

Weird ...

This little toys are, humm, cute. I need one.

Well. Back to work after some holidays. Will speak about it another day. I wrote this diary entry mostly for mathieu : Tu as jusqu'à 3 mois avant la fin de ton report pour demander un report pour CDI, donc jusqu'en maî 2001.

A propos, si quelqu'un à un truc pour changer de CDI et garder son report ...

Work : I am working on a Rich Media generation tool, an admin (a gfx co-worker who has clue) will produce some templates (SMIL and HTML -- in fact XHTML) and some keys/values stuff with the keys in the templates, and the luser will have a dynamic HTML form to fill. This will be a simple app for the moment, all the keys will be replaced by their value, and whereas the SMIL and XHTML are XML-based, I will not use DOM facilities, only the text struture.

In fact, I will do a very incremental process, because :

1. with a prototype and lots of releases, management will see that I am "working".
2. I setup a CVS and will use it.
3. I looked at XP and there seems to be good things in it, for example incremental process.

Too bad, I have no time to put a Junit test framework in place, so I will go live for this time.

I wrote the specs for the app yesterday, today I will do some HTML form workflow and UML diags of the app.

Ok. (Yes, real-time diary :p)

"La Défense" isn't cool. You can't open windows in these fscking buildings, and the offices are way too warm (I'm in T-shirt.).

And hey ... all the J2EE stuff are way too cool to build and to work with, all the classes, API and docs, but the deployement is a night-mare. I need to update my TODO list : find a solution, guy, trick ... cause I DON'T want to make deployement of my code. And I don't have time to. Need to speak about this to my boss ...

Ok.

A long time since I wrote an entry.

It's 02:25 PM here, just have a good lunch and drink with coworkers. I'm following a training for the week at "La Défense" in Paris. This is a sort of "place" with lots of buildings, and the weird thing is that there is people who are *living* here. It is not that bad to work, you come at 09:00 AM and you leave at 07:00 PM and voilà.

Yesterday, I bought a funny little white hat for my best friend, she'll be so cute with that :)

The training stuff is about java/JSP/Servlet/JavaBean/EJB ... (sorry you Sun people, I forgot the "TM" stuff) and it is not that bad to have a training, books are good to learn but sometimes training makes things go faster and easier to understand.

Weather is so hot in Paris for a December, there is no snow :( (unlike all the people on their diary : "Hey, we've got 40cm here", I hate you men :pp)

In fact I'm very tired.

Ok, not a lot of things to say and I have a bunch of people to certify.

My NetApp report isn't finished. Will try to finish it this afternoon.

My tests are almost finished, I will polish the report tomorrow, and I need to make the conclusions. There is a few things to know to use efficiently the filer, so I will add it. And I will have the opportunity to see some people from NetApp in Paris, they've got a demo lab.

Ok, just a little update :

1. the URL of the report is now http://xenon.medisite.net:8080/huge/report.html
2. one guy from the VAR which sells us our servers came this morning and helped me tuning the NetApp. It seems that there is a few tricks to know.
3. our switch *have* the trunking feature, which is cool :-) I just need to activate that on the switch and to configure the Quad Ethernet NIC on the filer.

Ok, one thing I remembered after posting my entry yesterday :

- In fact I was running *2* bonnie on the filer, the first on /vol/vol1 and the other on /vol/vol2, so the 2900 K/sec on per char / Sequential Output is just wrong. Quickly, I think I had to double that number. In fact, just forgot these

Here is the method I will use :

My test host (xenon) will be on a 10/100 switch (DLINK 3216) with the filer (using the built-in NIC). There is also 5 or 6 servers on the switch but I don't think it will interact. I will mount the filer /vol/vol1 with NFS (v2) and do a bonnie -2000 (not -1000 cause the filer is a 512 MB one). This first test will test the HD stuff and the network stuff of the filer, but I think the bottleneck is the network between my host and the filer. Yesterday I got 3/5 MB / sec traffic, it seems that bonnie generate a lot of traffic, maybe it is the data to be written in the file. I will look at the source later.

To test only the HD stuff, I think I will create a 10/20 GB file on one volume and mv it onto the other. I hope there will be the least possible networking stuff involved in this second test to stress the HD.

I also updated the report to add a SCSI 2 HD I have at home and a IDE HD.

And it seems that my lwm screenshot was a lot more popular than the bonnie report :

9:56 [huge:pts/2] xenon ~% grep screenshot.png /var/log/apache/access.log | wc -l
     26
9:56 [huge:pts/2] xenon ~% grep report /var/log/apache/access.log | wc -l
     3

I got my potato CDs Saturday and I installed the beast yesterday at home, no surprise, it is Debian. I'm running a woody at work, thanks to the large bandwith but at home I just have a little cnx, only IRC, little dls and a little web surfing. It will be cool to have a tested platform so I can work without do software install and worrying about that. I plan to test GTK developpment, GNOME libs and XML stuff, I have some ideas in mind so I will play with that. Unfortunaly, I don't have the source CDs but I can buy them at Ikarios and I will.

I also switched on lwm, E annoyed me. lwm is cool, although I made some "configuration" : put the cursor in black instead of red, the borders are 6 pixels wide, I use helvetica, I set the DEFAULT_TERMINAL to "term-custom" which is a alias on 'xterm -bg grey -fg black' so I can change that easily, and I swapped button 1 and 2 on the mouse. I also removed the "hostname:" in the title of the windows. The lack of "Alt-Tab" combo annoy me cause I must use the mouse to change windows and the lack of virtual desktops also. I will see if I can hack this one easily from wm2 and for the "Alt-Tab", it will be harder cause I never do X coding.

This afternoon, I tested our demo NetApp, I will put some results online when finished. I had some trouble cause the bottleneck is IMO the network system, not the disks system. There is the built-in 10/100 NIC and a Quad 10/100 NIC which support trunking or fail over. So a 4 * 100 Mbits virtual interface would be cool if we had a switch which support trunking. And we don't have one. The test procedure is : NFS mount, bonnie -s 1000 and look at results. I will say more on this tomorrow.

You will find the preliminary report here. There is also a screenshot of my lwm desktop.

I didn't put the NetApp benchmark but I had 2900 K/sec versus 5017 K/sec for my box. As sh pointed to me, "it sucks". But I will redo the test tomorrow.