Hi Domenic ! As i am the one who creates the gentoo ebuilds (which are still in testing mode as you know) i will try to help you with the ebuild install if you want.
The problem with the LD_ASSUME_KERNEL=2.2.5 is that the glib is not compiled on gentoo systems to support this. There is a bug in the gentoo bug system about this (see bugs.gentoo.org) and several entries in the discussion forums (forums.gentoo.org). So you will not get happy with trying to install the binary distribution on gentoo if your glib has been compiled to support NPTL. The thing that you will have to do in this case is to manually recompile your glibc again without NPTL and then use the LD_ASSUME_KERNEL. (see http://forums.gentoo.org/viewtopic.php?t=96278). If you want and need help to install the ebuild you can also contact me directly ore post a message in the gentoo forums. regards Hermann On Tue, 2004-06-22 at 06:09, Domenic Figliomeni wrote: > Hi, I am running gentoo linux with: > kernel version 2.4.23-aa2 > gcc-3.3.3-r6 > python-2.3.3-r1 > perl-5.8.2-r1 > bison-1.875 > > After failing to successfully install maxdb from sources using the > development ebuild (gentoo specific), due to an unknown dependency > problem, I opted for the binary release: > maxdb-all-linux-32bit-i386-7_7_00_12.tgz; md5 was ok > > I set up the user sdb and the group sdba as per default, and ran the > executable ./SDBINST. > I chose 10) All; installation exited abnormally > > Without too much debugging work here I decided to get the latest version > available: > maxdb-all-linux-32bit-i386-7_7_00_14.tgz; md5 was ok > I got the same problem at the same point. > > I then went through to test each profile with the following results: > 0) installation exited abnormally > 1) success > 2) success > 3) success > 4) success > 5) success > 6) success > 7) success > 8) success > 9) success > 10) installation exited abnormally > > Obviously, a problem with the server install. I watched each step and > the installer first tested all the tgz files that it would extract with > 100% success. When the second run (real extraction) occurred, the > installer successfully extracts the first three archives, namely: > SDBBAS.TGZ > SDBUTL.TGZ > SDBKRN.TGZ > > at which point I get: > > checking unpacked archive... ok > installation exited abnormally at (date...time..etc) > > I am expecting the next thing to do would be to unpack PCR7500.TGZ. > However, this doesn't even start. > > I manually extracted the archive PCR7500.TGZ and get: > PACKAGEDATA > instpcr > pcr.1st > runtime/7500/bin/irconf > runtime/7500/bin/irtrace > runtime/7500/lib/libpcr.so > > the md5's for each of the three runtime files matches those offered in > pcr.1st > > The following text is found inside PACKAGEDATA > > INTERFACE_VERSION = "0.1" > PACKAGE_NAME = "PCR 7500" > PACKAGE_VERSION = "0" > SOFTWARE_VERSION = "7.5.00.14" > MODE = "32" > MIN_VERSION = "7.4.02.00" > TYPE = "FULL" > TEST_FILE = "runtime/7500/lib/libpcr.so" > ALLOW_SKIP = "1" > MAGIC_MODE = "64TO32" > DEFAULT_USER = "sdb" > DEFAULT_GROUP = "sdba" > PART_OF = "Client,Server,Runtime For SAP AS,APO LiveCache,Development,C > Precompiler,DB Instance" > IS_TOP_OF = "Runtime For SAP AS" > REQUIRE = "Base >= 7.5.00.14" > FILELIST = "pcr.lst" > SCRIPT = "instpcr" > > If anybody else has seen this problem or has some idea about what is > going wrong or better still a solution, please reply (btw: > LD_ASSUME_KERNEL=2.2.5 gives me other errors about libc.so.6 and also > still get the same error as above). I suspect that possibly the > dependency problem I was experiencieng when building from source may be > related, so I have included versions of bison, perl, python and gcc (at > top). Is there anything else that maxdb server depends on, particularly > at the flawed part of the install? > > Regards, > Domenic Figliomeni > <ftp://ftp.sap.com/pub/maxdb/7.5.00/maxdb-webtools-linux-32bit-i386-7_5_00_12.tgz>
signature.asc
Description: This is a digitally signed message part
