@jacquelinepuimevallejo: #tensión

Jacqueline Puime Val
Jacqueline Puime Val
Open In TikTok:
Region: CH
Wednesday 07 October 2026 22:40:23 GMT
1032
106
6
7

Music

Download

Comments

kiras534
kiras534 :
esta sociedad no está en tensión, está podrida, no puedes decir nada en seguida te llaman de todo, y te desean lo peor del mundo
2026-10-08 01:32:36
3
ana.romn676
Ana Román :
estamos de imfarto👍👍👍👍👍👍
2026-10-08 08:02:51
1
edgardogarcia8130
edgardo12345567 :
Desde Uruguay quiero mandarles un fuerte abrazo a toda España y poder recordarles un viejo dicho que dice. Crea un problema y propone una solucion.
2026-10-08 14:22:44
1
corsa5190
corsa :
son los bots socio listos..
2026-10-08 13:07:28
1
eltiolavara6
eltiolavara :
😳😳👍👍
2026-10-08 09:49:43
1
username15an2
w2@ :
🤔
2026-10-08 18:55:59
0
To see more videos from user @jacquelinepuimevallejo, please go to the Tikwm homepage.

Other Videos

How do you ship a new version without taking your app offline? Blue-green deployment. Instead of one production environment, you run two. Blue is live — every customer is on it. Green is identical, sitting there with zero traffic. You deploy version 2 to green. Not on top of the thing your customers are using. Then you test it privately: can users log in, can they add to cart, does the checkout page reach the payment API, do the health checks pass? If something breaks — nobody notices. Your customers are still on blue. When green looks good, you don't deploy again. You move the traffic. The load balancer stops sending users to blue and starts sending them to green. Almost instantly, green is production. Five minutes later your monitoring shows checkout errors jumping from 0.4% to 18%? You don't rebuild version 1. You don't redeploy it. Blue is still sitting there with the last working release — so you flip the switch back. That's the real power here: rollback is a routing change, not a deployment. The part nobody warns you about is the database. Both environments share it. Rename a column in v2 and blue will crash the second you roll back — because it still expects the old name. That's why schema changes have to be backward compatible: add the new column, ship code that reads both, migrate the data, drop the old one later. Expand and contract. Two environments. One live, one ready. One switch between them. Save this for your next release 🔁 #devops #systemdesign #softwareengineering #backend #coding
How do you ship a new version without taking your app offline? Blue-green deployment. Instead of one production environment, you run two. Blue is live — every customer is on it. Green is identical, sitting there with zero traffic. You deploy version 2 to green. Not on top of the thing your customers are using. Then you test it privately: can users log in, can they add to cart, does the checkout page reach the payment API, do the health checks pass? If something breaks — nobody notices. Your customers are still on blue. When green looks good, you don't deploy again. You move the traffic. The load balancer stops sending users to blue and starts sending them to green. Almost instantly, green is production. Five minutes later your monitoring shows checkout errors jumping from 0.4% to 18%? You don't rebuild version 1. You don't redeploy it. Blue is still sitting there with the last working release — so you flip the switch back. That's the real power here: rollback is a routing change, not a deployment. The part nobody warns you about is the database. Both environments share it. Rename a column in v2 and blue will crash the second you roll back — because it still expects the old name. That's why schema changes have to be backward compatible: add the new column, ship code that reads both, migrate the data, drop the old one later. Expand and contract. Two environments. One live, one ready. One switch between them. Save this for your next release 🔁 #devops #systemdesign #softwareengineering #backend #coding

About