[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[LDM #TMA-441215]: Beta test 6.9?



Gilbert,

> Notes to myself:
> 
> Hmmm. Build environments tested. No 64 bit. Steve says it will work on 32
> bit. Hmmm, he hasn't tested it on CentOS 5.5, which is what wil happen
> here. I wonder if I smell a Monday coming on here. Hope this works.
> 
> Login environment...didn't have to change a thing. So far, so good.
> 
> I'm going to back up everything in my /home/ldm/etc directory.
> I don't want to lose anything. Just in case! ...done.
> 
> Huh. Weird unzip command. What's this pax command all about?

pax(1) is the UNIX standard unpacking utility. It replaces both tar(1) and 
cpio(1). It's capable of modifying file pathnames, which is used to put the 
source-files in the "ldm-6.9.0.4/src" directory.

> Dunno, but
> here we go.
> 
> gunzip -c ldm-6.9.0.4.tar.gz | pax -r '-s:/:/src/:'
> OK, that was fast.
> 
> cd /home/ldm/ldm-6.9.0.4/src
> 
> Ha! Love that config.guess file. Forecasting in winter is like that
> sometimes.
> 
> ./configure with all these options afterwards? Gonna just do it without
> the options and see where that gets me.
> 
> Asked me for root password, as warned. OK, done. That was fast.
> 
> Hmmm, changed my mind. Erased ldm-6.9.0.4, started over and used:
> 
> ./configure --localstatedir=/dev/shm

I take it you want the product-queue to be on a ramdisk.

> Now it's time to install.
> 
> make install >make.log 2>&1
> 
> Ambiguous output redirect.

You must be using a non-standard user-shell (e.g., csh(1), tcsh(1)). The 
instructions say to use the command "make install >&! make.log" if using csh(1) 
or tcsh(1).

> Well, that did nothing. Huh. OK, let me do it my way:
> 
> make install >/home/ldm/logs/install1.log
> 
> Now we're rolling. On screen, I see this:
> 
> /home/ldm% make install > /home/ldm/logs/install1.log
> parser.c: In function ÃxmlParseChunk__internal_aliasÃ:
> parser.c:11570: warning: comparison between signed and unsigned
> parser.c: In function ÃxmlParseInNodeContext__internal_aliasÃ:
> parser.c:12916: warning: pointer targets in passing argument 1 of
> ÃxmlFindCharEncodingHandler__internal_aliasà differ in signedness
> xmlIO.c: In function ÃxmlCheckHTTPInput__internal_aliasÃ:
> xmlIO.c:3680: warning: unused parameter ÃctxtÃ
> xmlmemory.c: In function ÃxmlMallocLoc__internal_aliasÃ:
> xmlmemory.c:208: warning: format Ã%ldà expects type Ãlong intÃ, but
> argument 4 has type Ãsize_tÃ
> xmlmemory.c: In function ÃxmlMallocAtomicLoc__internal_aliasÃ:
> xmlmemory.c:276: warning: format Ã%ldà expects type Ãlong intÃ, but
> argument 4 has type Ãsize_tÃ
> xmlmemory.c: In function ÃxmlReallocLoc__internal_aliasÃ:
> xmlmemory.c:352: warning: format Ã%ldà expects type Ãlong intÃ, but
> argument 4 has type Ãsize_tÃ
> xmlmemory.c:352: warning: format Ã%ldà expects type Ãlong intÃ, but
> argument 5 has type Ãsize_tÃ
> testapi.c: In function Ãgen_xmlValidCtxtPtrÃ:
> testapi.c:468: warning: unused parameter ÃnoÃ
> testapi.c: In function Ãdes_xmlValidCtxtPtrÃ:
> testapi.c:474: warning: unused parameter ÃvalÃ
> testapi.c: In function Ãtest_xmlBufferSetAllocationSchemeÃ:
> testapi.c:18828: warning: comparison of distinct pointer types lacks a
> cast
> testapi.c: At top level:
> testapi.c:315: warning: Ãgen_unsigned_longà defined but not used
> testapi.c:322: warning: Ãdes_unsigned_longà defined but not used
> testapi.c:327: warning: Ãgen_doubleà defined but not used
> testapi.c:336: warning: Ãdes_doubleà defined but not used
> testapi.c:341: warning: Ãgen_unsigned_long_ptrà defined but not used
> testapi.c:346: warning: Ãdes_unsigned_long_ptrà defined but not used
> testapi.c:390: warning: Ãgen_debug_FILE_ptrà defined but not used
> testapi.c:393: warning: Ãdes_debug_FILE_ptrà defined but not used
> testapi.c:763: warning: Ãdesret_doubleà defined but not used
> testapi.c:12199: warning: Ãgen_char_ptr_ptrà defined but not used
> testapi.c:12202: warning: Ãdes_char_ptr_ptrà defined but not used
> testapi.c:13137: warning: Ãgen_const_char_ptr_ptrà defined but not used
> testapi.c:13140: warning: Ãdes_const_char_ptr_ptrà defined but not used
> testapi.c:25914: warning: Ãgen_xmlElementContent_ptrà defined but not used
> testapi.c:25917: warning: Ãdes_xmlElementContent_ptrà defined but not used
> runsuite.c: In function ÃmainÃ:
> runsuite.c:1183: warning: control reaches end of non-void function
> runxmlconf.c: In function ÃmainÃ:
> runxmlconf.c:614: warning: unused parameter ÃargcÃ
> runxmlconf.c:616: warning: control reaches end of non-void function
> testWriter.c: In function ÃtestXmlwriterFilenameÃ:
> testWriter.c:163: warning: format Ã%010dà expects type ÃintÃ, but argument
> 4 has type Ãlong intÃ
> testWriter.c: In function ÃtestXmlwriterMemoryÃ:
> testWriter.c:442: warning: format Ã%010dà expects type ÃintÃ, but argument
> 4 has type Ãlong intÃ
> testWriter.c: In function ÃtestXmlwriterDocÃ:
> testWriter.c:712: warning: format Ã%010dà expects type ÃintÃ, but argument
> 4 has type Ãlong intÃ
> testWriter.c: In function ÃtestXmlwriterTreeÃ:
> testWriter.c:978: warning: format Ã%010dà expects type ÃintÃ, but argument
> 4 has type Ãlong intÃ
> /usr/bin/install: cannot stat `./*.html': No such file or directory
> make[4]: [install-data-local] Error 1 (ignored)
> ldm_xdr.c: In function Ãxdr_feedparÃ:
> ldm_xdr.c:235: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type
> ldm_xdr.c: In function Ãxdr_comingsoon_argsÃ:
> ldm_xdr.c:289: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type
> ldm_xdr.c: In function Ãxdr_datapktÃ:
> ldm_xdr.c:307: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type
> ldm_xdr.c: In function Ãxdr_datapktdÃ:
> ldm_xdr.c:326: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type
> ldm_xdr.c: In function Ãxdr_ldm_replytÃ:
> ldm_xdr.c:368: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type
> ldm_xdr.c:372: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type
> ldm_xdr.c:376: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type
> ldm_xdr.c:380: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type
> ldm_xdr.c: In function Ãxdr_fornme_reply_tÃ:
> ldm_xdr.c:428: warning: passing argument 2 of Ãxdr_referenceckà from
> incompatible pointer type

Yeah. There are certain warnings in the "libxml2" and XDR packages that I can't 
eliminate.

> LdmProxy.c: In function Ãlp_newÃ:
> LdmProxy.c:689: warning: assignment from incompatible pointer type
> pqcopy.c: In function ÃcopyProductÃ:
> pqcopy.c:70: warning: assignment discards qualifiers from pointer target
> type
> pqutil.c: In function ÃmainÃ:
> pqutil.c:1310: warning: initialization discards qualifiers from pointer
> target type

I think I can eliminate these warnings. Thanks.

> The next step will perform those installation actions that must be
> executed by the superuser.  If this step doesn't work (for example,
> if you don't know root's password), then the superuser will have to
> manually execute the command "make root-actions".
> 
> Enter root's password:
> 
> -----------------------------------------------------------------
> Typed in root's password, and then I was back out to prompt.
> 
> Configuring the ldm. blah blah, yes, I know all of this, because I have
> prior installation. But, new users will obviously need this.

I pray you're being facetious. :-)  Those new instructions are crucial.

> Oooh, metrics. What's that? Need to learn more. Will add it to crontab
> now. Like that new LDM check option. I'll set it for once every 30
> minutes. Wait. I use "tcsh" for LDM.

Unless your crontab(1) file is non-standard, the cron(8) daemon *should* use 
"sh(1)" to execute all entries. At least, that's what the standard demands.

> It says the LDM user must use bash.
> But, I have bash installed, so let's see if the crontabs still work
> anyway.
> 
> Huh. Says I have errors. Doesn't say what.. OK, try "/bin/bash" to the
> path for bash. Nope:
> 
> crontab: installing new crontab
> "/tmp/crontab.XXXXsmjmiU":21: bad day-of-week
> errors in crontab file, can't install.
> Do you want to retry the same edit?
> 
> That would be this line:
> 
> 2,32 * * * bash -l -c 'bin/ldmadmin check >/dev/null'
> 
> OK, weird. Commented out for now. Now the crontab is happy.

You need 5 fields before the command; you only have 4.

> Already have my own ldm start script. I'm good there.
> 
> OK, time to make the queue
> 
> ldmadmin mkqueue
> 
> Wait, that made the queue in /home/ldm/data/ldm.pq.
> I wanted it in /dev/shm!
> 
> OK, stopping now while I figure out why that happened,
> after I specified the option in the ./configure
> command.
> 
> Reverting to 6.8.1, for now. Awaiting Steve to answer this question.

If you install in a pre-existing, pre-6.9 LDM user account, then the configure 
script will notice that and use the existing "~/data" subdirectory for the 
product-queue (just like the pre-6.9 releases) rather than the value of the 
"localstatedir" variable ("~/var" by default). I thought I was being clever.

Check to see that's what happened by executing the command 
"~/ldm-6.9.0.4/bin/regutil | grep queue". I expect the pathname of the 
product-queue will start with "$HOME/data".

> *******************************************************************************
> Gilbert Sebenste                                                     ********
> (My opinions only!)                                                  ******
> Staff Meteorologist, Northern Illinois University                      ****
> E-mail: address@hidden                                  ***
> web: http://weather.admin.niu.edu                                      **
> *******************************************************************************

Regards,
Steve Emmerson

Ticket Details
===================
Ticket ID: TMA-441215
Department: Support LDM
Priority: Normal
Status: Closed


NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.