Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×
Bug

Salesforce Triggers 15-Hour Shutdown After Faulty Script Starts Granting View/Modify Access (zdnet.com) 29

Friday Salesforce "was forced to shut down large chunks of its infrastructure," ZDNet reports, calling it one of the company's biggest outages ever: At the heart of the outage was a change the company made to its production environment that broke access permission settings across organizations and gave employees access to all of their company's files. According to reports on Reddit, users didn't just get read access, but they also received write permissions, making it easy for malicious employees to steal or tamper with a company's data...

Salesforce said the script only impacted customers of Salesforce Pardot -- a business-to-business (B2B) marketing-focused CRM. However, out of an abundance of caution, the company decided to take down all other Salesforce services, for both current and former Pardot customers. "As a result, customers who were not affected may have also experienced service disruption, including customers using Marketing Cloud integrations," Salesforce said.

A status update at Salesforce.com reports that the final duration of the service disruption was 15 hours and 8 minutes.
This discussion has been archived. No new comments can be posted.

Salesforce Triggers 15-Hour Shutdown After Faulty Script Starts Granting View/Modify Access

Comments Filter:
  • by pegdhcp ( 1158827 ) on Sunday May 19, 2019 @06:44AM (#58617564)
    ...that despite being extremely difficult to customise and modify flow, rules per organisation, Salesforce has a serious market share. That much "complexity" is bound to collapse occasionally if rarely. This time it seems the problem is/was? very very global...
    • by Anonymous Coward

      ...that despite being extremely difficult to customise and modify flow, rules per organisation, Salesforce has a serious market share. That much "complexity" is bound to collapse occasionally if rarely.

      Oracle has a serious market share too, and their product is absolute garbage.

      It's both sad and pathetic, but quite often market share is a measure of mass ignorance, not quality.

      This time it seems the problem is/was? very very global...

      Do you know what a public company calls getting hacked in the 21st Century? A "faulty script". Don't bother asking what the problem was.
      You're not going to get the truth anyway. They have shareholders to think about.

      • by Anonymous Coward

        You must be a conspiracist. Oracle has more than one product, and you just do not like some of them. And many courageous companies will tell you the truth these days, but things are rarely basic and simple to explain, the readers also need to be able to read and have enough brain to understand...

        • by Anonymous Coward

          You must be a conspiracist. Oracle has more than one product, and you just do not like some of them. And many courageous companies will tell you the truth these days, but things are rarely basic and simple to explain, the readers also need to be able to read and have enough brain to understand...

          The public has been educated enough to understand "we got hacked", and it's hardly a conspiracy to find a publicly traded company bending the truth around a "faulty script" in order to protect the shareholder.

          You act as if we've never seen companies behave that way before. We've seen it countless times. In fact, we've seen it often enough to expect it.

      • by Anonymous Coward

        Its called production testing - live customers is just a bonus.
        Who approved that change, and why was there no quick rollback?
        For every time you are caught out, there are probably a dozen more incidents swept under the carpet.

      • by ranton ( 36917 )

        Do you know what a public company calls getting hacked in the 21st Century? A "faulty script". Don't bother asking what the problem was.

        You must have never went through the process of getting SOC 2 or 3 compliance if you think it would be easy for Salesforce to lie that drastically about the cause of this incident. Sweeping an incident no one knows about isn't as hard, but there will be regulators looking into this. The controls Salesforce would need to have in place to obtain SOC compliance would almost certainly not allow them to hide a security breach by calling it a faulty script. I've only had to deal with SOC 1 compliance in the past,

      • by Cederic ( 9623 )

        Oracle had one of the best database on the market for many years, and built a large customer base because of that.
        Oracle bought one of the top ERMs on the market, and acquired and built a large customer base because of that.
        Oracle bought the best CRM package on the market, and acquired and built a large customer base because of that.
        Oracle bought my favourite hardware provider, and totally fucked that one up.
        Oracle bought one of the two best application servers on the market, and acquired and built a large

    • ...that despite being extremely difficult to customise and modify flow, rules per organisation, Salesforce has a serious market share.

      Suddenly, without us ever noticing the company's growth, I notice that the tallest building in just about every major US city is the Salesforce Tower.

  • and auditing before committing.

    This is a beginners mistake... speaks volumes about the company.

    • This is why I don't commit to vendors without verifying what's running in the background. If I didn't explicitly ask for it, I don't want it.
      That is why sales force has intractable problems. I stopped using sales force for anything serious or non-serious when I kept getting the white screen of death saying numbers not found.

    • by ranton ( 36917 )

      clearly they needs some better change control and auditing before committing. This is a beginners mistake... speaks volumes about the company.

      It takes significant arrogance to believe anyone has perfect change control and auditing which makes mistakes like this impossible. There is always room for human error, and if you don't think there is room for error in your process it only shows you have hit the limit of your creativity and imagination. That isn't a strong criticism; everyone has a limit to their capabilities. And that limit is where the room for error comes from.

      One problem with applications as big as Salesforce is that the smallest appli

  • by Anonymous Coward

    A company I previously worked for started with their primary customer user interface being written in SalesForce sending data to our backend. SalesForce had a similar long outage and we had to scramble to put up something that at least our call centers could use to take orders by phone.

    There was a huge cry from our sale people and our merchants about the issue and we were getting constant bombardments for "when will it be back up". SalesForce will not commit to any SLAs on outages so we no matter how much

  • by LostMyAccount ( 5587552 ) on Sunday May 19, 2019 @08:25AM (#58617742)

    Even if video streaming doesn't work out for them, I feel like Netflix could pivot into systems engineering and consulting for the way they've built their environment for reliability and how they regularly test it by clobbering parts of it intentionally.

    So many other systems out there, even ones not as burdened with the usual sense of "legacy systems", aren't built that way from the beginning. They're a tangled mess of dependencies which make change testing and fault tolerance extremely difficult and seldom possible to test.

    • by Anonymous Coward

      Second this. However, it's easy to intentionally break things when you're the latest hotness. I'd love to have a chaos monkey running where I work, but we're too busy building new things and fixing old things. To also have stuff breaking randomly (yes, I know ChaosMonkey is opt-in) would just drive us up the wall.

      • The argument against this would simply be that this is because you (or your management) don't have (or want) the resources to make your companies engineering world class. It's not like Netflix doesn't have to build new things or fix old things. They simply hire and pay top of market for a number of world class developers that can sustain both existing and new products at a world class level. You really do get what you pay for, and each 9 in your SLA requires a LOT more work than the last.
    • by Cederic ( 9623 )

      Netflix have an incredibly simple environment compared to a company like Salesforce.

      They have full control over the content, the customer capabilities, their change processes and the logic they execute. You try asking Netflix to give you admin access over your own account, let you upload your own films, let you change the caching rules, the licensing agreements, the encoding algorithms.

      Netflix have done a great job but that doesn't mean for a moment that they could deliver something the scale and complexity

      • A fair criticism, but with something like Salesforce or many other cloud-based systems they introduce their own complexity to drive up licensing revenue, which in turn makes the system more complicated and more difficult to make reliable.

The Tao is like a glob pattern: used but never used up. It is like the extern void: filled with infinite possibilities.

Working...