5 steps to rebuild S8700 duplication server.
How to turn a bad situation into a good experience
By the end of the ‘5 steps to rebuild S8700 duplication server’ blog post, you will understand the core techniques that I recently used to help one of our clients, by restoring their VoIP solution, and turned a horrific event into a better experience.
I was recently sent to fix a defective S8700, which wasn’t able to save translation whenever the active server happened to be the S8700 Secondary server. This particular customer experienced a power surge causing major damage to Server B’s Motherboard.
Because Server A was up and running fine, the customer didn’t feel the urgency of swapping the bad server. It is during moments like this that you have to be thorough with your clients and explain how severe the problem might be, even though in the surface everything may look fine.
These are the steps=
- 1.- Engage with your with peers
- 2.- Be prepared
- 3.- Develop a plan, and introduce it
- 4.- Get to work
- 5.- What to do when things don’t work out
1.- Engage with your peers
Once the project or service request gets assigned to you, the first step that you need to take is to gather all previous work performed for this client that is related to this incident. You have to reach out to each individual engineer and listen which tasks they have completed successfully and which ones have failed. You want to concentrate on the tasks that failed. This will help you pinpoint where the problem lies.
2.- Be prepared
Once you understand the scope of the work, it is time to get ready by downloading the Software ISO, Service Packs, and Patches available for this particular Server. The Server was running 2.2.2, and the support.avaya.com website didn’t have the same patches that I needed. PLDS is another option for you to find Patches and any other type of software related to your customers. And if you get lucky, you might be able to find these in the server itself.
3.- Develop a plan, and introduce it
At the arrival to the site, scope the site and meet with your contact(s). Listen first, then draw a plan. Your plan should include Backups, screenshots, software and patching installation procedures, restore backups, Interchange servers, and finally a test plan.
4.- Get to work
Now that the customer is aware of your scope of work, he/she will appreciate that you allowed them into your world by sharing your ideas. Always be open and flexible to change your steps if the customer requires or suggests it.
Backup the existing system configuration to local and external media such as HDDs, and SD Card Readers, as well as an FTP directory, if available. Besides, the system configuration you should also backup the License and Authentication files. When the backup status is completed, verify that the XLN, OS, Security Files were backed up.
In theory, you shouldn’t have to go through the backup procedures, when working with Duplicated or HA Servers. The Secondary server should pull all of the files and configuration needed from the Primary. I am suggesting doing the backups as a precaution.
Screenshots – If you are using a Windows computer, you will most likely have access to the Snipping-Tool (see resources), and MS-Word. Open both of these and connect to the Active Server, in my case was server A. Once connected, navigate using the SMI through the Maintenance/Server-Management window, select Configure individual services.
Using the Snipping-Tool screen grab the Server Identities, Interfaces, Ethernet Adjuncts, * External DNS Server Config, Network Time Server, **Set Modem Interface
DAL Boards – These have to be the same on both servers. Run the command ‘cat/proc/mdd’ from the CM’s Shell CLI.
Server Reload – Restart the server and load the ISO, service packs, patches, and restore backups. You have to install each patch the same way the Primary Server had it (them) installed. Don’t load the License or Authentication files, these will be transferred over when you run the save translation command from the Active server A.
Swapping the server – Mark or take a picture of the cables connected in the back of both Server A and Server B, then remove any media attached to the bad server. (Modems and Card-readers). Proceed by replacing the physical server, connect everything back, and start the server.
*- Enabled DHCP Service on this server for IPSIs was selected. This is very important, if left uncheck the existing IPSIs won’t be able to function.
** – For B Servers increase the PPP IP Address assigned to the modem by two Octets.
5.- What to do when things don’t workout
Every situation will be different and things sometimes won’t work out as planned. These are some techniques to help you:
Server doesn’t see the CD-ROM Drive – Each component gets mounted to the Linux kernel, failing to do so will cause similar issues. The faster way to fix this symptom is by replacing the CD-ROM Drive or Server. For more advanced technicians, you can mount the drive, then try to kickstart the installation. And lastly be sure to have a good burned copy of the software by checking the MD5 Checksum, and the correct disk.
Servers are not shadowing – The main cause is DAL board issues or mismatch.
Can’t save translation from the B Active server – Patches were loaded wrong.
Resources
Blog Post – My Server Crashed Now What?
Please note: I reserve the right to delete comments that are offensive or off-topic.