2007 April 26 : 7:01 pm
315 F Hostel, NIT Calicut.
Thadiyan left ten minutes back.
"It was great four years with all you guys....", he was repeating...
All of us here: Eoeo,Sri Raj, Aju and me went to see him off in front of the hostel. Some of the others are going to city along with him....
The room next to me is now vacant. Mine will be vacant too, in a matter of days.
Selwin has already left... vacated his room and gone. ( Actually i surrendered the room for him.) That was on wednesday morning. Rajeen and Subin went with him too, but they will be back.
Now it is raining here.... another one of the rains in nitc.... feels a bit gloomy though....
Thursday, April 26, 2007
Wednesday, April 25, 2007
The Two Admins
This is a tribute to one of the most dedicated and passionate guys i've ever seen - a tribute to his unique nature and undying spirit which i discovered during my short time of working with him.
The administratorship of Athena was something i always wanted since the early days of second year. Getting the password of Athena was actually a dream come true for me. Sreedal was a good friend of mine even before that; but it was the knowledge of this password that brought us closer and made us big friends.
Our initial duties as admins were to read all the READMEs Luko had prepared and stored root's home and reading the logwatch mails for Athena. Logwatch would report all the failed login attempts, disk space statistics etc. ( My personal favourite being the failed attempts to become superusers - it was interesting to know who all were trying su and sudo and how many times at a stretch). I was a bit confused about what all things to do on an everyday basis. Backing up stuff was automated, thanks to our seniors.
Things were fine till the end of s6, since luko was there.
Then we had the full responsibility of the lab. The lab was shifted to the new lab complex and several stuff were done right from scratch. The NAT machine became Thallettan's ( hmm... thats what we call Sreedal, with respect :) !! ) pet. Interestingly, Vinod sir had assigned NAT to Sujith, but it was thallettan who did everything on it. Incidently, this caused sir to give NAT charge to Athena admins themselves in our junior year. He was very attached to this machine - may be because of the sheer complexity it offered due to all its network settings ( iptables, proxy settings etc) and the power of this single machine doing all the authentication and port forwarding stuff to every single request going out and coming in of the software lab. I remember two things with regard to this:
Our daily routine became modulated with the software lab. We would stay there together and go for food together. But one day he was playing with the NAT because of some problem and when i called him for lunch, he told me to go and he would have food later. I went had my food, went to the hostel, slept a bit, had my bath , had my dinner and as usual went back to the lab. He was still there and guess what... he didn't have his lunch, tea or dinner... why? the NAT was still having problems. I tried my best to make him go and have his dinner, but he won't budge. His dinner that day was the usual MC food we have after closing the lab at 12...
Another thing was regarding the password of the NAT. Eventhough we have full freedom to change any passwords in the lab, we have to keep Jerry and Ajith informed about this stuff. A new lab for the MCA was setup and they took the network connection from our lab. They were having network problems and when Noble wanted to see some settings in the NAT, Jerry gave him NAT's password. ( That isn't the way things should work. Jerry should have logged on into set system, showed stuff to Noble and should have logged off.) Sreedal came to know of it and he was really furious. Noble getting to know our password wasn't a big deal at all - he's such a responsible person. Sreedal would have no probs about it. But Jerry should have followed the norms.
And guess what? NAT got a new password. A very jumbled cryptic confusing password with a lot of numbers and about 25-30 letters long. I tried several times to make it by-heart. No success still. For a pretty long time, Sreedal was only one who knew the password. Whenever i had to check NAT ( obviously that happened only when Sreedal was not around) i had to call him and he would dictate the password letter by letter through phone!!
Sreeedal always wanted to learn new things. He was helping a lot of people around with tech stuff. He was a frequent invitee to the CNC to troubleshoot stuff. However busy he may be, he would never let down anyone asking for help. And he would take all the responsibilities around. He was co-ordinating the activities of CSEA as well. This taking up all the responsibilities around became so intense that i wouldn't to go to the dept with him for fear of landing into more responsibilities.
Another great time was the starting of the eighth sem. We didn't have classes for about a month. We used this time to reinstall the machines in the lab. Fedora Core 6 had released around that time and we decided to use that. We were looking forward to install beryl in all the machines in the lab. I was like mad about the awesome features of beryl. It took us more than a week to do it. But we did it. There were a lot of issues we had to resolve. Lots of machines were in a mess. There were extra users in some of the systems. Backups in some etc.
It was one of the most happy moments at the lab. To see a long array of machines with the fedora 6 login screens on them.
After that things were pretty smooth. Only challenges were the new web server coming up and conducting the Programming contest.
The CSEA programming contest was an interesting challenge. Sreedal was very passionate about it. He was trying to create some frame work for the online programming contest from scratch. I suggested checking sourceforge. We found some frameworks for it and chose one for our use. It was a good piece of software but the configuration stuff was pretty huge. It was having a lot of dependencies ( another common challenge in linux system administration - dependencies) like mySQL, php, java vm etc. Thallettan spend over a weeks time to get everything up and running. Unfortunately, i didn't sit with him in setting up stuff because i was busy with my project. On the day of the competition, he had to go home. And i was left with a highly complex system of which i had no idea of. I was hoping things to go smooth since i had no clue how to troubleshoot if something went wrong.
Unfortunately, something did go wrong. A phone based troubleshooting session with him didn't work. The event had to be postponed. I was pretty desp about it.
Anyway we conducted it pretty well the next week. It was interesting to see a lot of people turning up for the prelims.
Sreedal, this is for you and all those wonderful days we spend in the software lab.
The administratorship of Athena was something i always wanted since the early days of second year. Getting the password of Athena was actually a dream come true for me. Sreedal was a good friend of mine even before that; but it was the knowledge of this password that brought us closer and made us big friends.
Our initial duties as admins were to read all the READMEs Luko had prepared and stored root's home and reading the logwatch mails for Athena. Logwatch would report all the failed login attempts, disk space statistics etc. ( My personal favourite being the failed attempts to become superusers - it was interesting to know who all were trying su and sudo and how many times at a stretch). I was a bit confused about what all things to do on an everyday basis. Backing up stuff was automated, thanks to our seniors.
Things were fine till the end of s6, since luko was there.
Then we had the full responsibility of the lab. The lab was shifted to the new lab complex and several stuff were done right from scratch. The NAT machine became Thallettan's ( hmm... thats what we call Sreedal, with respect :) !! ) pet. Interestingly, Vinod sir had assigned NAT to Sujith, but it was thallettan who did everything on it. Incidently, this caused sir to give NAT charge to Athena admins themselves in our junior year. He was very attached to this machine - may be because of the sheer complexity it offered due to all its network settings ( iptables, proxy settings etc) and the power of this single machine doing all the authentication and port forwarding stuff to every single request going out and coming in of the software lab. I remember two things with regard to this:
Our daily routine became modulated with the software lab. We would stay there together and go for food together. But one day he was playing with the NAT because of some problem and when i called him for lunch, he told me to go and he would have food later. I went had my food, went to the hostel, slept a bit, had my bath , had my dinner and as usual went back to the lab. He was still there and guess what... he didn't have his lunch, tea or dinner... why? the NAT was still having problems. I tried my best to make him go and have his dinner, but he won't budge. His dinner that day was the usual MC food we have after closing the lab at 12...
Another thing was regarding the password of the NAT. Eventhough we have full freedom to change any passwords in the lab, we have to keep Jerry and Ajith informed about this stuff. A new lab for the MCA was setup and they took the network connection from our lab. They were having network problems and when Noble wanted to see some settings in the NAT, Jerry gave him NAT's password. ( That isn't the way things should work. Jerry should have logged on into set system, showed stuff to Noble and should have logged off.) Sreedal came to know of it and he was really furious. Noble getting to know our password wasn't a big deal at all - he's such a responsible person. Sreedal would have no probs about it. But Jerry should have followed the norms.
And guess what? NAT got a new password. A very jumbled cryptic confusing password with a lot of numbers and about 25-30 letters long. I tried several times to make it by-heart. No success still. For a pretty long time, Sreedal was only one who knew the password. Whenever i had to check NAT ( obviously that happened only when Sreedal was not around) i had to call him and he would dictate the password letter by letter through phone!!
Sreeedal always wanted to learn new things. He was helping a lot of people around with tech stuff. He was a frequent invitee to the CNC to troubleshoot stuff. However busy he may be, he would never let down anyone asking for help. And he would take all the responsibilities around. He was co-ordinating the activities of CSEA as well. This taking up all the responsibilities around became so intense that i wouldn't to go to the dept with him for fear of landing into more responsibilities.
Another great time was the starting of the eighth sem. We didn't have classes for about a month. We used this time to reinstall the machines in the lab. Fedora Core 6 had released around that time and we decided to use that. We were looking forward to install beryl in all the machines in the lab. I was like mad about the awesome features of beryl. It took us more than a week to do it. But we did it. There were a lot of issues we had to resolve. Lots of machines were in a mess. There were extra users in some of the systems. Backups in some etc.
It was one of the most happy moments at the lab. To see a long array of machines with the fedora 6 login screens on them.
After that things were pretty smooth. Only challenges were the new web server coming up and conducting the Programming contest.
The CSEA programming contest was an interesting challenge. Sreedal was very passionate about it. He was trying to create some frame work for the online programming contest from scratch. I suggested checking sourceforge. We found some frameworks for it and chose one for our use. It was a good piece of software but the configuration stuff was pretty huge. It was having a lot of dependencies ( another common challenge in linux system administration - dependencies) like mySQL, php, java vm etc. Thallettan spend over a weeks time to get everything up and running. Unfortunately, i didn't sit with him in setting up stuff because i was busy with my project. On the day of the competition, he had to go home. And i was left with a highly complex system of which i had no idea of. I was hoping things to go smooth since i had no clue how to troubleshoot if something went wrong.
Unfortunately, something did go wrong. A phone based troubleshooting session with him didn't work. The event had to be postponed. I was pretty desp about it.
Anyway we conducted it pretty well the next week. It was interesting to see a lot of people turning up for the prelims.
The collaboration in our final year projects was another significant part. Sreedal did the image processing part of our groups project. He created an algorithm for detecting roads in Matlab. He experimented on it for a pretty long time making it pretty elegant.
An interesting thing about him is the fact that even though he would do things with passion, he would not take the same interest in presenting it. I would say the presentation layer would be desp most of the times. That must be the reason why he would not be the class topper even though he was the one with the strongest concepts there. Same thing happened with our project stuff too.
He did it in Matlab and left it there for about a month. I had to ask him umpteen number of time s before i could get the VC++ version from him. I virtually had to make him sit down before our project machine and stand guard to make him convert it!!
Another notable period was the FOSS meet. We had to take up the responsibility for all the technical stuff for the FOSS event. The initial enthusiasm began to wane as the things to be done started to mount up. Just one week before the start there were so many things to be done and no people in our team other than the two of us. We had to meet all the requirements of the speakers and even create questions and manage the overnight programming contest. Amarjith wanted to conduct it in very special ways. We first thought of doing it in proper ACM ICPC style and were planning to host a local server thing. But Amarjith wanted some other parameters for evaluation. To be frank, both of us did not like the plan, particularly because of our experiences with programming contests in the ICPC style.
Anyway, with the help of a few night-outs, we got the things done. We faced a lot of problems during the meet too. The speakers wanted wireless net in the venues and the guest house. All e had is a Apple Airport which could be configured only from a mac. Sreedal had to work pretty hard to get the net up and running. Config had to be done from the lab ( our mac is there ) and had to be taken to the 'site' for checking. He was running around for the whole of Day one for this messing up all our plans for the programming contest for day 2 night. In true nitc spirit, we did it just in time, with the help of Eohan and Sri Raj and a dedicated bunch of juniors. The programming contest night was awesome. I took our project laptop to the lab and was working on it, checking the solutions submitted occassionally. Sreedal and eoeo were around and the lab was all crowded with 'programmers'. We were pretty tired by 9 pm (thanks to the night-outs in the previous two days) but were enjoying the stuff as the night progressed. It was over by 4:30 - 5:00 in the morning and we went back to the hostel to catch some sleep before things started again by 8:00.
By Feb-March, we handed over the lab charge to our juniors. Sreedal, by that time had created a web server for the department. Andomeda was taken from the lab and a spare machne ( he started calling it Venus , the name's stuck, i think) was brought in. It runs ubuntu coz yours truly was pretty fascinated with the new ubuntu at that time :). That was probably the last official stuff we did in the lab.
Sreedal was selected as the Best Outgoing B.Tech Student from the department. He was the obvious choice : always ready to take up any responsibility, passionate about what he does, first to grasp any new concept, studious, intelligent, simple, helpful....
Sreedal was ranked 12th in India in the GATE 2007 and he's gonna join IISc for his masters. His flair for theoretical stuff and deep insight shall surely lead him to the higher realms of computer science.
Thalletto, thanks for all those wonderful days. All the best da.
Thursday, January 25, 2007
The Fellowship of Athena
It's gonna be one year now.
The image of Deepak Lukose writing down Athena's password for us on the back side of my notebook is still fresh in my memory. We were the new admins for Athena; new admins for the software lab. I still remember the first time i logged into Athena as 'root', the first time i used 'sudo' and the first time i read the Logwatch mails for admin. I still cherish the nice feeling looking out into the night through the C hostel window after i threw that password-paper, torn into pieces.
And two days back Vinod sir asked us to find new admins for Athena.
It actually took some time to dawn on me. We are going to give Athena's password to someone. 192.168.40.99 is gonna be someone else's.
The one year of essentially 'ruling' the lab was really an unforgettable period of my life. On the average we spend more than 6 hours per day in the lab. Sreedal started the habit of spending the evenings there first. I followed in a week.
One of the first jobs we had was setting up the lab in the new building. In fact the dept had shifted the comps and had done the networking part by then. We started with the NAT machine. I mean Sreedal started with the nat... i wasn't interested in that machine and still isn't. He actually reconfigured it from scratch. Really nice job it was.
Then came the job of creating user accounts for the second years'. Athena already had the scripts for creating mass accounts. It was actually to create accounts based on a basename and with numbers appended, like gamma19. That's why we were gammas and our seniors were deltas. I modified the script to use names from a file. And the juniors got their roll numbers as their usernames.
The fact that Athena runs on a 64 bit version of Red Hat Enterprise Linux (RHEL) usually make installing things difficult. Source compilations usually get stuck at points like checking the system type and getting the shared libraries. RPM are difficult to find for most of the packages. So the news that we had to install things like Nachos java and Ocaml weren't greeted with enthusiasm. I managed to install the java version of Nachos but the version was pretty bad in itself. Documentation was very poor and ultimately the lab used the cpp version itself. In fact we had to start another server "Nachos Server" (192.168.40.40) with a 32 bit version of FC5 for the lab. We actually ported the user accounts from Athena and Andromeda (with the help of the andro admins) to the new server.
Ocaml was much tougher. I fact we couldn't get it done properly. Luko came to the rescue. He set up Ocaml in time from his home. Yeah, he did it from home. The details can't be disclosed!!
From that point onwards things were much smoother. We found ways to get around many problems. Prolog installation turned out to be difficult but i found a set of static binaries for 64 bit system. That's why Athena has 'Yet Another Prolog' installed rather than gprolog or SWI-Prolog.
One of the things i always wanted to do in Athena was to change the /etc/motd. I had always wanted to greet the users with ascii pics. I actually did a lot of changes.
This was the first screen i set up:
After a few weeks i came up with something like this :
And after some time i settled down with this:
Now Athena is still going on, but its high time that its reinstalled. We asked permission to do it but we were asked to postpone it to the end of March. I hope that would be a befitting goodbye to Athena from two people who cared for her for a whole year.
The image of Deepak Lukose writing down Athena's password for us on the back side of my notebook is still fresh in my memory. We were the new admins for Athena; new admins for the software lab. I still remember the first time i logged into Athena as 'root', the first time i used 'sudo' and the first time i read the Logwatch mails for admin. I still cherish the nice feeling looking out into the night through the C hostel window after i threw that password-paper, torn into pieces.
And two days back Vinod sir asked us to find new admins for Athena.
It actually took some time to dawn on me. We are going to give Athena's password to someone. 192.168.40.99 is gonna be someone else's.
The one year of essentially 'ruling' the lab was really an unforgettable period of my life. On the average we spend more than 6 hours per day in the lab. Sreedal started the habit of spending the evenings there first. I followed in a week.
One of the first jobs we had was setting up the lab in the new building. In fact the dept had shifted the comps and had done the networking part by then. We started with the NAT machine. I mean Sreedal started with the nat... i wasn't interested in that machine and still isn't. He actually reconfigured it from scratch. Really nice job it was.
Then came the job of creating user accounts for the second years'. Athena already had the scripts for creating mass accounts. It was actually to create accounts based on a basename and with numbers appended, like gamma19. That's why we were gammas and our seniors were deltas. I modified the script to use names from a file. And the juniors got their roll numbers as their usernames.
The fact that Athena runs on a 64 bit version of Red Hat Enterprise Linux (RHEL) usually make installing things difficult. Source compilations usually get stuck at points like checking the system type and getting the shared libraries. RPM are difficult to find for most of the packages. So the news that we had to install things like Nachos java and Ocaml weren't greeted with enthusiasm. I managed to install the java version of Nachos but the version was pretty bad in itself. Documentation was very poor and ultimately the lab used the cpp version itself. In fact we had to start another server "Nachos Server" (192.168.40.40) with a 32 bit version of FC5 for the lab. We actually ported the user accounts from Athena and Andromeda (with the help of the andro admins) to the new server.
Ocaml was much tougher. I fact we couldn't get it done properly. Luko came to the rescue. He set up Ocaml in time from his home. Yeah, he did it from home. The details can't be disclosed!!
From that point onwards things were much smoother. We found ways to get around many problems. Prolog installation turned out to be difficult but i found a set of static binaries for 64 bit system. That's why Athena has 'Yet Another Prolog' installed rather than gprolog or SWI-Prolog.
One of the things i always wanted to do in Athena was to change the /etc/motd. I had always wanted to greet the users with ascii pics. I actually did a lot of changes.
This was the first screen i set up:
After a few weeks i came up with something like this :
And after some time i settled down with this:
Now Athena is still going on, but its high time that its reinstalled. We asked permission to do it but we were asked to postpone it to the end of March. I hope that would be a befitting goodbye to Athena from two people who cared for her for a whole year.
Subscribe to:
Posts (Atom)
