2008 is a year full of sorrows. A lot of people are suffering from the Economic Tsunami caused by the losing credibility of the financial system in the U.S due to the sub-prime crisis. The negative effect of this man made disaster is likely to continue in the coming few years. Fortunately many countries are showing positive attitude to work among all of them together to overcome this problem as soon as possible. I hope the level of confidence will restore next year, which is year 2009. I wish everyone a good year of 2009!
2008年，对于人类，尤其中国人来说，是蛮多灾多难的。最重要的世界大事莫过于起源于美国次贷危机的金融海啸了！至今各国的对金融海啸的反应都非常的积极，我觉得主要是信心崩塌的问题。相信当每个人的信心渐渐恢复了之后，全球将重新迅速迈入了又一个欣欣向荣的阶段。最重要的是各国都能把这次的事件当作反面教材，牢牢记得，如此错误千万莫重犯！最后，期待经济能于2009年复苏！人人皆能分得糖果！
Wednesday, December 31, 2008
Tuesday, December 23, 2008
Production
Finally my WBI adapter for Kenan (a finance system built by Accenture) has gone for PRODUCTION last week. I thought after that everything will be fine. However...
I hit the Tuxedo connection error again. The murderer is the Tuxedo client library again! This time i migrated the jar version that i edited and put to the one at production server. It did not work! I was panicked!
Luckily i deployed the system earlier than the agreed time and i discovered the problem earlier. I still had the whole night trying to fix the problem. I went back and worked from home until 1 something in the morning. Finally i got the solution which is i have to set TUXEDO relevant environment variables at either my startup script or my user's profile (.profile). (It's AIX machine)
I went to bed after that. 4am in the morning i was awakened by a call from maxis staff to stand by for testing before production. What frustrated me most is a lot of tables and database objects are not granted with enough privileges. This problem has been troubling me till the morning.
Also, a lot of late requirements came out only after the users found out some data integrity issue. That even worsened the problem.
The fight continued until 5 something in the late afternoon next day. The client decided to switch to a backup plan. Thus the production was announced failed.
I hit the Tuxedo connection error again. The murderer is the Tuxedo client library again! This time i migrated the jar version that i edited and put to the one at production server. It did not work! I was panicked!
Luckily i deployed the system earlier than the agreed time and i discovered the problem earlier. I still had the whole night trying to fix the problem. I went back and worked from home until 1 something in the morning. Finally i got the solution which is i have to set TUXEDO relevant environment variables at either my startup script or my user's profile (.profile). (It's AIX machine)
I went to bed after that. 4am in the morning i was awakened by a call from maxis staff to stand by for testing before production. What frustrated me most is a lot of tables and database objects are not granted with enough privileges. This problem has been troubling me till the morning.
Also, a lot of late requirements came out only after the users found out some data integrity issue. That even worsened the problem.
The fight continued until 5 something in the late afternoon next day. The client decided to switch to a backup plan. Thus the production was announced failed.
Subscribe to:
Posts (Atom)