BIND 10 #1292: Xfrin retransfer shared object "sqlite3_ds.so" not found
BIND 10 Development
do-not-reply at isc.org
Wed Nov 23 19:32:47 UTC 2011
#1292: Xfrin retransfer shared object "sqlite3_ds.so" not found
-------------------------------------+-------------------------------------
Reporter: jreed | Owner: jinmei
Type: | Status: reviewing
defect | Milestone:
Priority: | Sprint-20111206
critical | Resolution:
Component: | Sensitive: 0
Unclassified | Sub-Project: DNS
Keywords: | Estimated Difficulty: 4
Defect Severity: High | Total Hours: 0
Feature Depending on Ticket: |
Add Hours to Ticket: 5 |
Internal?: 0 |
-------------------------------------+-------------------------------------
Changes (by jelte):
* owner: jelte => jinmei
* status: reopened => reviewing
Comment:
(I am assigning this directly back to you jinmei, since you had the
original request and proposal in the first place, feel free to make
unassigned if you don't have time before tomorrow, or throw chairs in my
direction if you hate the idea too much)
Okay, I have a proposal for a slightly different middle-term solution to
this problem than the one you sent to bind10-dev; there are no additional
values passed, but the DataSourceClient container now calls a function
that makes an absolute path for dlopen() to use;
- if the file does not end with .so, _ds.so is appended (so that once we
do have configuration, any name can be used for home-grown modules)
- if the file is not absolute, it is made absolute. By default this uses
$(prefix)/lib (but we can change to whatever directory we want to put our
modules in), unless B10_FROM_BUILD is set in the environment (at this
moment I can't think of any way not to need something like that somewhere
in the code path; we'll always have to be able to override the value for
in-tree runs and tests), then it will use
$(abs_top_builddir)/src/datasrc/.libs.
I have an initial implementation of this in trac1292_2; please take a look
if you have the opportunity.
This branch also has a slight addition to the lettuce test; it prints
bindctl stdout/stderr if bindctl calls fails when sending a command. Also,
a number of tests now need to set B10_FROM_BUILD so that the install
prefix isn't used when using the factory loader.
BTW, I have not checked, but if this works, i suspect it solves several
other related tickets as well (#1385 and #1421, maybe more)
--
Ticket URL: <http://bind10.isc.org/ticket/1292#comment:20>
BIND 10 Development <http://bind10.isc.org>
BIND 10 Development
More information about the bind10-tickets
mailing list