forked from flibertigibet/SRS
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Interview
107 lines (84 loc) · 9.26 KB
/
Interview
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
DA-IICT Interview Summary
=========================
## System : Dashboard
## Project Reference : Software Engineering Course (IT314)
## Participants: Interview of- Nimish Patel (Head of System Administration of DA-IICT)
## Interview taken by- Saloni Jain, Barkha Bhojak
## Date: 2/2/2016 Time : 3:30 pm
## Place: Lab Building, Staff Room
## Duration of the interview: 90 minutes
## Purpose of Interview : The interview is taken to gather basic information about current internet working model in Da-iict campus.
#### Inferences of the interview:
1. Role as System Administration head
* New deployment
* Managing overall system
* Maintenance and upgradation of IT infrastructure on whole.
2. Hostel problems and steps taken to solve them:
* Individual problem: it includes improper internet functioning in the system of any individual hostelite.
* Net connectivity issues in a wing.
* Steps taken to solve above problems:
+ Firstly a person is sent to that particular hostel room where internet connectivity issues exist.
+ Person checks for minor problems in the local machine(laptops) like DNS service problem and physical connectivity problem.
+ Physical connectivity is checked by checking if the lan and the cable is working properly. If yes, then the connectivity from hostel room to hostel server room of that particular wing is checked.
+ In the server room, there are switches consisting of 24(or 48) ports for connecting rooms port to switch.
+ On the switch, the port on which problem is coming is placed in the neighbouring port just to check if that particular port is not working or that switch on whole.
+ If that particular port is not working, then it is replaced with a temporary port else the switch is replaced.
3. The internet budget of DA-IICT is divided into 2 types:
* OPEX: Operational budget. I t includes the operational cost or the running cost of the already existing system. It includes basic cost like maintenance cost, software renewal cost.
* CAPAX: it stands for capital expenditure. It includes deployment of new software in the existing system, also the upgradation of the existing system. For example, the recent cost paid to get campus wise MATLAB license. The request for CAPAX is made either at the end of the year or before the beginning of the new semester.
4. Upgradation is conducted from time to time:
* Desktops and servers are upgraded from time to time. Last upgradation was done 5-8 years back.
* There is also upgradation of information in the beginning and end of each semester. It includes activation of new accounts for the students getting admission in July- August. Also the deactivation of cyberoam of the students completed their final year.
5. Lab problems mainly include:
* Physical connectivity problem.
* PCs not shutdown properly or rebooting properly.
* PC getting slow down or crash happening. This kind of problem is rarely encountered.
* Internet connectivity problem. This kind of problem is also hardly encountered.
* There is a person who is hired to keep a check on lab systems. His job is to go on a check for the above stated problems once the official lab timings are over.
6. The internet service is being provided by BSNL ISP in the campus. The bandwidth being provided ranges from 300-400 Mbps. Very often bandwidth problems are encountered when bandwidth gets reduced significantly due to one of the following reasons:
Heavy load due to increased number of users. It generally happens in night or in case of some event where average number of active users grows from 600-800 to 100-1200.
Or there is virus in some PC or system which causes drop of real packets along with bad packets , leading to loss of packets , therefore information and resulting in poor bandwidth.
7. Working of the help desk and system administration in order to ensure proper internet and server working in the campus:
+ The helpdesk is comprised of 6 people who belong to some third party:
* 2 people are from network and server farm having complete technical knowledge.
* 3 people are proper field engineers. There job is to go to hostel rooms , labs, faculty block or any place in the campus where an issue is detected.
* 1 person is assigned to coordinate calls. He is the person in charge of receiving calls and recording the complaint and telling it to the appropriate person so that problem can be solved.
+ The ideal working hours of help desk is from 8.30am to 10pm. The time from 8.30 am to 5 pm is considered as the most active time, therefore all 6 person responsible are present in the help desk. From 5pm to 9pm, there are 2 persons in charge. And from 9pm to 10 pm, there is only one person in charge.
+ The problems are identified through continuous monitoring of emails, intranet, webmail, websites , hostel main switch, wing wise switch and many more. This monitoring is done 24*7. During monitoring if some issue is detected , an alert or email is sent to help desk automatically. Within one minute of issue detection, the mail or notification is sent.
+ The problem with the current system is that if a student finds that his/her internet is not working properly after 10pm , then notification may or may not be sent automatically. In most such cases, student has to notify the help desk. Even if an alert is sent then also such issues are resolved next morning only. This is because help desk has prioritized the issues. If the problem is found to be genuine or big (like in case of internet not working in the whole wing), then only help desk gets active after 10pm. It first tries to solve that problem remotely.
+ The limitations in solving problem after 10pm is because of unavailability of appropriate person.
+ In general, it takes 4-5 hours to take backup of the whole system information. So incase of power cutoffs, help desk coordinates with electric team and asks them to shift power source to DG(inverters) 2 to 3 hours prior to electric cut off.
8. Recent upgradations of the system and continuous upgradations are:
* In 2013, the ground floor line was upgraded. Also 350 PCs were incorporated in the new labs made for Computer Science students.
* There were only 2 rooms in Resource center where wifi was present 2 to 3 back. But now whole RC, lab building, cep and cafe have a good speed and range wifi.
* Every year 5-6 switches are being bought and replaced.
* In the month of February, 2016 there is the proposal of replacing the core and core backup switch.
* Also, campus has spent a good amount in getting licenses for Dreamspark, Adobe, Matlab and many more.
* Also there is a proposal on converting lans in hostel into wifi. But this is highly dependent on budget.
9. Following are the servers :
+ Dedicated email server.
+ Dedicated cyberoam login for staff and students.
+ Firewall- Network authentication server. These 3 servers are virtual servers and are being designed using blade technology( where several switches are made on one single board. Blade provides extra storage as all switches have a common storage).
+ There are 5 web servers. Moodle and intranet comes under this.
+ Print server.
+ Online admission server.
+ Course submissions server.
+ NIS- server for fedora domain login in labs. This is meant for students only.
+ NAS- server for centralized storage. This is for faculty( example for lecture folders).
+ 4 Research servers are also in process for faculty, phd and M.Tech students. They were proposed by Research faculty (Amit Bhatt). The idea behind this is to have softwares already present in servers; students can submit code and then check it there only without requiring softwares individually. This will be quite handy and all will be able to work together.
+ Computation setup server for CS students. A new lab is made where 4-5 servers work as a cluster to do computing needed for CS.
10. User statistics of Da-iict user community:
* The peak number of users vary from 800-1000 per day. And average number of users are 700 per day. Similar data is being gathered from last 3 years. During lecture timing , i.e, from 8am to 1pm , the average number of users is approximately 450 which is very surprising.
* The same data 4-5 years back was very different. Average users use to be 300-350 with a peak of 450 users per day.
11. How does college block the websites?
* Cyberoam has categorized the url(s) , apps and websites according to their application.
* Most of the redirection websites are blocked at present.
There are 2 methodologies available for blocking.
+ Block All- In this all websites are blocked initially and then websites are unblocked or made active depending on the usage and requirement of system and its users.
+ Allow All- In this all websites are allowed initially and then certain websites are blocked. This is the method deployed in our college because it is easy to sort unwanted websites instead of sorting important or required websites. In this blocking can be done in many ways. One such method is block by starting word(example: blocking torrent automatically blocks torrentpower)
12. The authorities having the right to change, add, delete or update any information on intranet are:
+ UG convener
+ PG convener
+ Dean Academics
+ Registrar
13. Lecture folder has read only rights for students.