The Database Administrator should have a morning checklist to reference each morning when going through their environment. This list below can be used as a tool, with this tool going through each section. We can then check them off and feel a little more at ease that the systems have been reviewed.
Let us get started. Some questions might come up when you first start in a Database administration role.
- Where do I start, what to review first thing in the morning?
- Once I start to review what the process I should follow?
- Do we have any notifications, calls, emails for an issue?
After years of experience most database administrators have a good idea in their mind what they need to check or start what to fix first. But if you have a list written down in a document or some type of reference.
Some examples through my years of experience.
- Most common a word document with a numbered list.
- Excel following the same concept as the word document.
- I have seen some tie a check list to SSRS reports then check off each report they have created.
- There are many different approaches to fit your style.
This list also depends on your environment. How many Database administrators do you have. Even more, depending if they are local, remote or even global.
Below is an example of how I have set up our environment that can touch on all those questions. Since we use team’s for communication in our environment. Also, have 4 onsite DBAs and other DBAs globally. I have created a team’s channel, then added files to that channel that have a standard morning checklist as the daily template. Then the on-call database Administrator for that day can reference this list and post it to the team’s channel once completed. With this process, everyone on the team and even globally is aware that if any issues they are being addressed. Another good reason this approach is good is that if any other DBA has information about an issue listed, they can respond to the posted checklist in teams for everyone to see.
Here we go. Let’s get start with the check list. This is not a rule or a standard just some suggestions to help the regular DBA have a process for daily checks.
Here we go. Let’s get started with the checklist. This is not a rule or a standard just some suggestions to help the regular DBA have a process for daily checks.
- Backups
- Verify the backup or backups have run successfully. Either manually or setting up some kind of alert to review. (Alerts will be covered in another blog)
- If a back up failed review the job history and try to identify the issue.
- Things to check duration of back up
- If the duration has a big variance. (Contact Network explaining your findings)
- If on a network and use file share. Check access if granted.
- If you have more than one database, verify which ones got back up and the once that weren’t.
- File share of Drive has significant enough space.
- Depending on the error you will need to research
- Disk Space
- Having one or more servers or locally. Depending on your set up you will need to check free space. If you see a variance, you will need to research the issue and resolve.
- Things to check duration of back up
- Sometimes times there can be an open transaction causing the log to fill up. (Logging space for reference will be in another blog)
- Job failures
- Multiple ways to check your job failures.
- Set up alerts to review the job failures using tools or your method you choose.
- Set up email notification\texts on failure to review
- Multiple ways to check your job failures.
- Right click on the failed job in SSMS and view the job history.
- SQL Log or Operating System checks(Windows Events)
- Check the SQL logs for errors
- If you are on windows, you can also review the windows events for the prior day
- Depending on the issue, normally it’s the application events
- If you have identified something unusual contact your networking team to review and fix.
- Server Performance
- This depends, do you have a monitory tool to review? Things to review long running processes, blocking or deadlocks. If you don’t have tools, you can set up processes to capture these.
- Depending on the issue, normally it’s the application events
- Server connectivity
- Ping is a good way to ensure you can connect to the server or server in question
- Replication (if you have this in your environment)
- Open Replication monitor to view
- Connect to the distributor.
- View the subscriptions
- You can adjust your list to your environment since everyone’s environment is different.
- Most use different methods and have different tools, scripts etc.
- Connect to the distributor.
- Open Replication monitor to view
- Backups
Conclusion.
I have listed out just some standard morning daily checks from my experience. Use this list as a reference for your environment. Good luck.
Discover more from SQLyard
Subscribe to get the latest posts sent to your email.
