September 18, 2015 at 4:47 pm
Hi I wanted to ask what do 'other' organizations do from an infrastructure and workstation setup look like for staff
that develop Java code in IDE on Linux but connects to a SQL Server.
E.g. How does local development work for developers?
(do teams local databases?)
(do teams use a shared dev database?)
I've seen 3 styles (worked 2/3 styles)
1) Developers have a local database and some subset of 'data' for testing code - then integrate db changes to "Dev" database
2) Developers all use a single common dev database for dev testing .. this though anticipates data doesn't change much and only code does.. or DB is refreshed nightly..
3) Developers each have their own DEV copy on a DevServer so that no one steps on each others toes.
Any ideas/suggestions? I'm trying to streamline the dev process/support to get teams to use MS SQL... but most teams are running a local virtual box or VM to for a local SQL Server DB and trying to develop code locally then push SQL code and App code out... seems very cumbersome.
thanks for any feedback
--------------------------------------------------
...0.05 points per day since registration... slowly crawl up to 1 pt per day hopefully 😀
September 19, 2015 at 9:19 pm
sqlsurfing (9/18/2015)
Hi I wanted to ask what do 'other' organizations do from an infrastructure and workstation setup look like for staffthat develop Java code in IDE on Linux but connects to a SQL Server.
E.g. How does local development work for developers?
(do teams local databases?)
(do teams use a shared dev database?)
I've seen 3 styles (worked 2/3 styles)
1) Developers have a local database and some subset of 'data' for testing code - then integrate db changes to "Dev" database
2) Developers all use a single common dev database for dev testing .. this though anticipates data doesn't change much and only code does.. or DB is refreshed nightly..
3) Developers each have their own DEV copy on a DevServer so that no one steps on each others toes.
Any ideas/suggestions? I'm trying to streamline the dev process/support to get teams to use MS SQL... but most teams are running a local virtual box or VM to for a local SQL Server DB and trying to develop code locally then push SQL code and App code out... seems very cumbersome.
thanks for any feedback
My take:
3 is ideal, using a representative copy (maybe even on representative hardware) that can be "refreshed" from prod on demand by the devs.
1 is OK as long as you follow-through with integration into a representative environment prior to shipping to QA.
2 is what I mostly see due to cost barriers and compliance or organizational constraints and can work OK however can be quite limiting to the dev team.
There are no special teachers of virtue, because virtue is taught by the whole community.
--Plato
September 20, 2015 at 11:35 am
Quick thought, I've used free tds, http://www.freetds.org/ for years, many of my workstations are Linux/Unix and freetds provides the equivilent of sqlcmd, works quite well (most of the time:-D). Does expand the options for development
😎
October 20, 2015 at 1:12 pm
Orlando Colamatteo (9/19/2015)
sqlsurfing (9/18/2015)
Hi I wanted to ask what do 'other' organizations do from an infrastructure and workstation setup look like for staffthat develop Java code in IDE on Linux but connects to a SQL Server.
E.g. How does local development work for developers?
(do teams local databases?)
(do teams use a shared dev database?)
I've seen 3 styles (worked 2/3 styles)
1) Developers have a local database and some subset of 'data' for testing code - then integrate db changes to "Dev" database
2) Developers all use a single common dev database for dev testing .. this though anticipates data doesn't change much and only code does.. or DB is refreshed nightly..
3) Developers each have their own DEV copy on a DevServer so that no one steps on each others toes.
Any ideas/suggestions? I'm trying to streamline the dev process/support to get teams to use MS SQL... but most teams are running a local virtual box or VM to for a local SQL Server DB and trying to develop code locally then push SQL code and App code out... seems very cumbersome.
thanks for any feedback
My take:
3 is ideal, using a representative copy (maybe even on representative hardware) that can be "refreshed" from prod on demand by the devs.
1 is OK as long as you follow-through with integration into a representative environment prior to shipping to QA.
2 is what I mostly see due to cost barriers and compliance or organizational constraints and can work OK however can be quite limiting to the dev team.
Thanks Orlando and Eirikur Eiriksson! (apologies for late reply but wanted to say thanks for reply)
I've offered freetds and provided instructions, hopefully that will help in the future.
It turns out I will need to perform a 'sales pitch' for lack of a better word, as the dev teams use Linux. I work for small company now and I need to come up with standards for most things such as default db for relation db development.
I must provide enough of an incentive. I'm a SQL Server DBA based on experience, but I also need to use open source DBs mysql, postgresql It has been a "developers' choice in terms of DBs since there was no DBA. There are that are developers indifferent / flexible and those resistant to change... due to development experience with open source only or a typical bias.. "unix is cool - microsoft sucks" probably.
Developers can use virtual machines to develop on SQL Server locally as well, but I need to make the installation/deployment to their linux laptops easy and seamless. Along with providing scripts to create SQL data/db. I have heard vagrant is a good tool to help seamless deployment. (not explored yet... Im not a linux guy :ermm:
--------------------------------------------------
...0.05 points per day since registration... slowly crawl up to 1 pt per day hopefully 😀
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply