Re: Does remote caching open up any security vulnerabilities?
Bill Deegan <[email protected]>
| Newsgroups | gmane.comp.programming.tools.scons.user |
|---|---|
| Message-ID | <CAEyG4CE93idns-C5B50PPSsxXHYZc9xqg9PBKNujCizFKvp9Mg@mail.gmail.com> |
Don, If you're seriously concerned (as in this is actually happening ever) about malicious alteration of your cachedir in your development environment, then you have lots of issues.. That's what I was referring to. Nope. Remote caching definitely has been asked for and there's a PR with an implementation as well. But as yet not merged. That predates the customizable cachedir implementation which dmoody mentioned. Indeed a SCons cachedir server, and plugins for (I think it's) meson's cachedir server would be great additions. -Bill On Thu, Sep 22, 2022 at 10:38 PM Don Baldwin <[email protected]> wrote: > Thanks for the quick response Bill. When you say we’ll have “lots of > issues to contend with”, what exactly are you referring to? Is Remote > Caching generally thought to be more of a headache than it’s worth? > > > > Thanks, > > Don > > > > > > *From:* Scons-users <[email protected]> * On Behalf Of *Bill > Deegan > *Sent:* Thursday, September 22, 2022 12:15 PM > *To:* SCons users mailing list <[email protected]> > *Subject:* Re: [Scons-users] Does remote caching open up any security > vulnerabilities? > > > > *WARNING:* This email originated from outside of Qualcomm. Please be wary > of any links or attachments, and do not enable macros. > > > > > > On Thu, Sep 22, 2022 at 12:03 PM Don Baldwin <[email protected]> > wrote: > > Hi, > > > > What precautions are in place to prevent someone from modifying a cached > file to inject nefarious code into a product? > > > > Currently there are none. > > If you're building software in such an environment, you have lots of > issues to contend with in addition to someone altering the cachedir files. > > > > Generally we don't check target file modification either (assuming nothing > depends on a given target file as source to another builder). > > Though in that case it would just case a rebuild of the target which > depends on it, not a specific notice that it had been modified. > > > > _Bill > _______________________________________________ > Scons-users mailing list > [email protected] > https://pairlist4.pair.net/mailman/listinfo/scons-users > _______________________________________________ Scons-users mailing list [email protected] https://pairlist4.pair.net/mailman/listinfo/scons-users