Re: 2.5.STABLE2 around the corner

On Wed, 12 Feb 2003, Henrik Nordstrom wrote:

> >From what I can tell everything is now in place for a 2.5.STABLE2
> release, and the current 2.5.STABLE1-20020212 snapshot is to be regarded
> as a 2.5.STABLE2 release candidate.
>
> If you can please give this a try to make sure there is nothing stupid
> which really should be fixed before we roll 2.5.STABLE2.

I'm keeping track of some minor bugs/changes that I would like to
make to 2.5 before O'Reilly publishes the squid book. I'm willing
to make the fixes, but am worried about breaking things (i.e., the
way people expect things to be) in the stable versions. They are:

CARP

        squid-2.5 CARP configuration/implementation is different
        than squid-2.x head. For example, 2.5 uses 'carp-load-factor='
        but 2.6 uses 'weight='

annoying bug in Number of fields per header distribution

        Bug counts 0 fields per header when calling httpHeaderClean()
        for new headers, before they are even used:

        id #flds count %total
         1 0 13363088 49.75
         2 1 18216 0.07
         3 2 979 0.00
         4 3 19 0.00
         5 4 1140906 4.25
         6 5 1241749 4.62
         7 6 168336 0.63
         8 7 442458 1.65
         9 8 859209 3.20
        10 9 1958607 7.29
        11 10 3070119 11.43
        12 11 2823685 10.51

ICP/HTCP treated differently. In some places, ICP and HTCP are
treated as equivalent. For example, in "Peer Selection Algorithm"
stats, "ICP" also counts HTCP. But in other places, HTCP is not
also counted. For example:

        - "Number of ICP messages received"
        - "Number of ICP messages sent"
        - "Cache Client List" stats
        - "List of Unknown sites sending ICP messages"
        - ...cacheClientIcpRequests.A.B.C.D in SNMP

syscalls.disk.unlinks confusion

        The counter is incremented for:
        - unlinks performed by main squid process
        - unlinks performed by DISKD processes
        The counter is not incremented for:
        - unlinks performed by AUFS threads
        - unlinks performed by the unlinkd process

cachemgr "Cache Utilization" page

        this page gives too much detail to be useful to most
        users, and the name sucks. it should be removed.

in storeCheckCachable(), no.release_request always 0

        because storeReleaseRequest() clears ENTRY_CACHABLE, and
        such objects get counted as no.not_entry_cachable instead

FETCHES under "Cache Client List"

        the FETCHES number counts requests forwarded for any reason
        (ICP, HTCP, Cache Digests, default parent).
        Following the FETCHES count is a percentage, based on the
        number of ICP replies. This percentage may be larger
        than 100. Just remove the percentage, or only count
        fetches due to ICP?

DISKD Q1, Q2 confusion

        documentation does not match the implementation. Tried to
        convince henrik of this before, but failed.

Redirectors recieve URIs with whitespace

        if "uri_whitespace allow" then redirectors may receive
        URIs with whitespace and not properly parse them.

IpcacheStats and FqdncacheStats

        I removed some members of IpcacheStats and FqdncacheStats
        in squid-2 HEAD, but did not MFC to squid-2.5 (or squid-3)
Received on Wed Feb 12 2003 - 12:07:16 MST