look some more technical detail needed, you may get some basic http cache background information from the SDK guide.
from that Admin Guide, just remember: User URL -> Maped URL -> MD5 key -> Disk from the SDK guide and codes, you man find out the very down side storage interface, which is far complex for a short mail to explain. please find more information in the SDK guide, to get some basic knowing about the IO interface. And you may need to dig into the codes to get your question resolved. you may find some information in Chinese here: http://zymlinux.net/trafficserver/ you can find out my contact information there too, feel free to contact me off list. PS: TrafficServer is a big monster, the first Chines guy who finnaly get conquest the codes, have spend about 6 months, in full time. hopes that will help you. 在 2011-03-11五的 20:42 +0000，ChangCheng写道： > > > > > John,Thanks your clearly explan from the point of view of the cache , for now > , i am confused about how TS start lookup cache. For example, when user > request www.google.com . follow this diagram. we can see ,the TS start cache > lookup after Os DNS lookup,The question is, what variable that TS used to > look up in cache for www.google.come, in other word, how does TS know if > there existed this page? i would like to know the process details about how > TS find if there existed this page? > Date: Fri, 11 Mar 2011 09:29:19 -0800 > > Subject: Re: How TS find object from object database (cache) > > From: jplev...@acm.org > > To: dev@trafficserver.apache.org > > > > I will answer from the point of view of the cache and perhaps someone else > > could answer the more detailed HTTP/1.1 > > questions which are outside of my expertise. > > > > The cache get's a hash (MD5) (of the URL after some massage I suppose, don't > > quote me on that). > > It looks it up in a big hash table which gives it 0 or more possible > > location for the head of the > > object on disk (the fragment containing the vector of headers and perhaps > > the body of one > > of the alternates). > > > > If it is a "miss" it tells the HTTP state machine. > > > > If it has a possible hit it it the HTTP state machine grabs the vector of > > alternates and > > compares it to the request to see if there is a match. If there is, it > > requests that object > > (which might already be in the fragment that was retrieved). It then > > streams that > > document to the user. For multiple fragment documents things are a bit > > more complicated. > > For documents with alternates and situations where the header is updated but > > not the contents > > things are more complicated as well. > > > > If you want pointers or have more specific questions I'll try to help you. > > I you want to > > know about HTTP stuff, perhaps someone else can help. > > > > Some details of the directory structure were just covered here (see an > > earlier post), perhaps > > in the archive if you were not subscribed. > > > > john > > > > 2011/3/11 ChangCheng <ccai...@msn.com> > > > > > > > > Hi guys:I'm glad to join this group, i'm currently a college senior > > > student > > > with a little extra bit of time on my hands and wanted to get a little > > > experience with TS.Currently,I have accrued some basic knowledge through > > > go > > > through TS Administrator's Guide.For now, there is a problem i can't > > > understand, How Ts find object from object database (cache) ? in page : > > > http://trafficserver.apache.org/docs/v2/admin/http.htm , step2 : it > > > mentioned Using the object address, Traffic Server tries to locate the > > > requested object in its object database (cache).. but i am not very sure > > > what is exctly object address, it is url or hash key for url? For example, > > > if user request www.google.com , when the TS receive request, how it find > > > this page from database? what is key for ts check page from database? i > > > really would like to konw the details about how TS find object from cache. > > > Thanks > >
