Because we decided a long time ago to keep positive as well as negative
entries in the cache as apps that were issuing repetitive requests could
do either negative or positive operations (not a great explanation but
one that tries to capture that the issue was on both sides of the fence).
Atleast that's what I recall as we did that a very long time ago.
On Fri, 11 Jun 2021, Brian P Bockelman wrote:
> I don't think it's incompatible with HTTP -- it's simply having problems with the sequence of operations FTS is performing. I believe they are something like this:
>
> ```
> mkdir /foo/bar/baz # error - does not exist
> mkdir /foo/bar # success
> mkdir /foo/bar/baz # success
> stat /foo/bar/baz # FAIL: the prior mkdir failed with "does not exist"
> ```
>
> Actually - @abh3, why doesn't the `mkdir` invalidate the negative cache entry in this case?
>
> --
> You are receiving this because you were mentioned.
> Reply to this email directly or view it on GitHub:
> https://github.com/xrootd/xrootd/issues/1466#issuecomment-859980522
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub, or unsubscribe.
[
{
"@context": "http://schema.org",
"@type": "EmailMessage",
"potentialAction": {
"@type": "ViewAction",
"target": "https://github.com/xrootd/xrootd/issues/1466#issuecomment-859981689",
"url": "https://github.com/xrootd/xrootd/issues/1466#issuecomment-859981689",
"name": "View Issue"
},
"description": "View this Issue on GitHub",
"publisher": {
"@type": "Organization",
"name": "GitHub",
"url": "https://github.com"
}
}
]
Use REPLY-ALL to reply to list
To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1
