I set up a redirector (fermiclooud039.fnal.gov) in front of a cache (fermicloud022.fnal.gov Config of Redirector ``` all.role manager all.manager fermicloud039.fnal.gov 1095 http.listingdeny yes all.export / if exec xrootd xrd.protocol http:8000 libXrdHttp.so xrd.port 1094 fi # Added so http would not redir to https all.export / xrd.allow host * sec.protocol host cms.dfs limit 0 lookup distrib mdhold 0 redirect immed retries 1 cms.fxhold noloc 15m 4h http.desthttps no http.trace all -debug xrootd.trace emsg login stall redirect xrd.trace conn cms.trace defer files redirect stage ``` Config of cache: ``` all.export / xrd.allow host * sec.protocol host # A lot of stale http connections from caltech #xrd.timeout idle 10m xrootd.trace emsg login stall xrd.trace all ofs.trace all http.trace all ofs.osslib libXrdPss.so pss.origin redirector.osgstorage.org:1094 pss.cachelib libXrdFileCache.so #pss.setopt DebugLevel 1 # Increase debug level pss.setopt DebugLevel 3 pss.trace all oss.localroot /tmp all.manager fermicloud039.fnal.gov:1095 all.role server pfc.trace debug **if exec xrootd xrd.port 1094 xrd.protocol http:8000 libXrdHttp.so fi** sec.protbind * none all.sitename Fermicloud-edgar-test ``` Notice that the xroot port and the http port of the cache do not match. What happens is that even with a http call the redirector redirects the cleint to the xrootd port instead of the http port: ``` [root@fermicloud022 stashcache]# curl -L -H "Cache-Control: max-age=86400" -v http://fermicloud039.fnal.gov:8000/user/dweitzel/public/blast/queries/query1 -O % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent Left Speed 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0* About to connect() to fermicloud039.fnal.gov port 8000 (#0) * Trying 131.225.154.173... * Connected to fermicloud039.fnal.gov (131.225.154.173) port 8000 (#0) > GET /user/dweitzel/public/blast/queries/query1 HTTP/1.1 > User-Agent: curl/7.29.0 > Host: fermicloud039.fnal.gov:8000 > Accept: */* > Cache-Control: max-age=86400 > < HTTP/1.1 302 Redirect < Connection: Keep-Alive < Content-Length: 0 < Location: http://fermicloud022.fnal.gov:1094/user/dweitzel/public/blast/queries/query1 < 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 * Connection #0 to host fermicloud039.fnal.gov left intact * Issue another request to this URL: 'http://fermicloud022.fnal.gov:1094/user/dweitzel/public/blast/queries/query1' * About to connect() to fermicloud022.fnal.gov port 1094 (#1) * Trying 131.225.154.156... * Connected to fermicloud022.fnal.gov (131.225.154.156) port 1094 (#1) > GET /user/dweitzel/public/blast/queries/query1 HTTP/1.1 > User-Agent: curl/7.29.0 > Host: fermicloud022.fnal.gov:1094 > Accept: */* > Cache-Control: max-age=86400 > * Recv failure: Connection reset by peer 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0 * Closing connection 1 curl: (56) Recv failure: Connection reset by peer ``` When it should be redirecting it to port 8000 not port 1094. I am tagging @djw8605 and @DrDaveD. Since there is a parent CVMFS issue https://sft.its.cern.ch/jira/browse/CVM-1822 -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/xrootd/xrootd/issues/1087 ######################################################################## 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