hns - 日記自動生成システム - Version 2.19.9

先月 2003年05月 来月
1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31
Namazu for hns による簡易全文検索
詳しくは 詳細指定/ヘルプを参照して下さい
検索式:

2003年05月31日() 旧暦 [n年日記]

#1 [NetBSD] 1.6T

/usr/mdec/installboot:

# pwd
/mnt/usr/mdec
# ./installboot -v /mnt/boot bootxx /dev/rwd0a
installboot: open raw: No such file or directory
# ls -l
total 552
-r--r--r-- 1 root wheel 2271 May 29 12:20 bootxx
-r-xr-xr-x 1 root wheel 97864 May 29 12:20 installboot
-r--r--r-- 1 root wheel 512 May 29 12:15 mbr
-r--r--r-- 1 root wheel 55400 May 29 12:20 ofwboot
-r--r--r-- 1 root wheel 59348 May 29 12:20 ofwboot.elf
-r--r--r-- 1 root wheel 55632 May 29 12:20 ofwboot.xcf
-r--r--r-- 1 root wheel 8192 May 29 12:20 ustarboot

#2 ap2-php4

... sapi_apache2.c:31: apr_strings.h: No such file or directory
In file included from sapi_apache2.c:32:
/usr/pkgsrc/www/ap2-php4/work/.buildlink/include/httpd/ap_config.h:58: apr.h: No such file or directory
/usr/pkg/include/apr-0/apr.h にあるのが見えていない。


2003年05月30日(金) 旧暦 [n年日記]

#1 [機械] 219 ?? IWsa 0:29.33 /usr/pkg/sbin/named

これって止っている ?
ttyp0:makoto@st3240  18:03:22/030530(~)> ps ax|grep named
 219 ?? IWsa 0:29.33 /usr/pkg/sbin/named 
kill -1 219 しても反応しない

1.6R:

ttyp0:root@st3240 18:09:36/030530(/home/makoto)# uname -a NetBSD st3240 1.6R NetBSD 1.6R (L2CR) #0: Sat May 17 20:04:39 JST 2003 makoto@st3240:/usr/s/src/sys/arch/macppc/compile/L2CR macppc

BIND 9.2.2:

ttyp0:root@st3240 18:09:44/030530(/home/makoto)# named -version さふいへば、こんな字が残っていた

/netbsd: sa_switch(219.10 flag 900004): Not enough stacks.:

May 30 11:34:12 st3240 /netbsd: sa_switch(219.10 flag 900004): Not enough stacks. May 30 11:34:12 st3240 /netbsd: sa_switch(219.10 flag 900004): Not enough stacks

また来た:

May 30 20:51:32 st3240 named[3941]: couldn't add command channel ::1#953: file n
sa_switch(3941.10 flag 900004): Not enough stacks.
sa_switch(3941.10 flag 900004): Not enough stacks.
May 30 21:01:44 st3240 /netbsd: sa_switch(3941.10 flag 900004): Not enough stack

netbsd-bugs: kern/20245:

core 有:

ttyp1:root@st3240 21:07:19/030530(/etc/named)# gdb /usr/pkg/sbin/named named.core
GNU gdb 5.0nb1
(gdb) where
#0 0x41b55ef4 in pthread__switch () from /usr/lib/libpthread.so.0
#1 0x41b55ee8 in pthread__switch () from /usr/lib/libpthread.so.0
(gdb)

kern_sa.c:

464 sa_switch(struct lwp *l, int type)
465 {
...  
505                 /*
506                  * XXX We need to allocate the sadata_upcall structure here,
507                  * XXX since we can't sleep while waiting for memory inside
508                  * XXX sa_upcall().  It would be nice if we could safely
509                  * XXX allocate the sadata_upcall structure on the stack, here.
510                  */
511
512                 if (sa->sa_nstacks == 0) {
513 #ifdef DIAGNOSTIC
514                         printf("sa_switch(%d.%d flag %x): Not enough stacks.\n",
515                             p->p_pid, l->l_lid, l->l_flag);
516 #endif
517                         goto sa_upcall_failed;

sysctl -a:

まさかこういう簡単な話 ?
> sysctl -a | grep stacks
proc.curproc.rlimit.stacksize.soft = 2097152
proc.curproc.rlimit.stacksize.hard = 33554432
# sysctl -w proc.curproc.rlimit.stacksize.soft=4194304
proc.curproc.rlimit.stacksize.soft: 2097152 -> 4194304
どこかで memory leak のようなことが起きていて、それがどこかが問題ということか。

#2 [MTA] send-pr したメールが送られていないなぁ

ということがあって、echo test | mail hoge@foo とかしても同じ。

/var/log/maillog:

May 30 18:34:28 harry sendmail[557]: h4U9YSrw000557: to=hoge@foo, ctladdr=makoto (2029/500), delay=00:00:00, xdelay=00:00:0 0, mailer=relay, pri=30030, relay=[127.0.0.1] [127.0.0.1], dsn=4.0.0, stat=Deferred: Connection refused by [127.0.0.1]
なんて書いてある。そうか sendmail を起動しておく必要があるのかと思って

/var/spool/clientmqueue:

May 30 18:34:28 harry sendmail[557]: dangerous permissions=40775 on queue directory /var/spool/clientmqueue/ と言っているので他の機械を見て、
drwxrwx--- 2 smmsp smmsp 1024 May 30 17:11 .
に直したら、その瞬間に起動したような気配。 で、/var/spool/clientmqueue に溜っているのを出すのはどうやるのかな というのがきょうの疑問
if [ -x /usr/sbin/sendmail ]; then
echo "Starting sendmail MTA daemon: /usr/sbin/sendmail -L sm-mta -bd -q25m"
/usr/sbin/sendmail -L sm-mta -bd -q25m
echo "Starting sendmail MSP queue runner: /usr/sbin/sendmail -L sm-msp-queue -Ac -q25m"
/usr/sbin/sendmail -L sm-msp-queue -Ac -q25m
fi
この辺かな。一つ目が MTA で、これが起動し終ってから二つ目を実行する。(& を付けてはいけない)

#3 [MTA] mail back to myself (Mx problem)

またこれで 38 通くらい失ってしまった。 内部用 dns


2003年05月29日(木) 旧暦 [n年日記]

#1 [network] ipnat/rdr/smtp

ipnat の rdr を使って smtp port を飛ばしていると、 EOM で完了しない相手が 1% くらいある。(と思う)
そこで 「mx で一度それ(rdr)が不要のところまで集めて、 そこから違う mx 内容を見せておいて内部に転送する」 という正統的な方法に変更。

しかし sendmail-8.9.3 + CF-3.7 だと:

「mx が違うところを指しているはずなのに、自分で受取って しまう」ような気がして、/etc/aliases で明示的に転送。
多分自分の設定のどこかがおかしいのだろう。

#2 [httpd] 2.0.46

5/30 日までに入替えた方がいいよ。とのこと。

#3 [NetBSD] build.sh

一応完了
==> Summary of results:
build.sh command: ./build.sh -D /export/20030516-utc/d -R /export/20030516-utc/r release
build.sh started: Thu May 29 13:30:56 UTC 2003
Bootstrapping nbmake
MACHINE: macppc
MACHINE_ARCH: powerpc
TOOLDIR path: /export/20030516-utc/src/tools/obj/tools.NetBSD-1.6R-powerpc
DESTDIR path: /export/20030516-utc/d
RELEASEDIR path: /export/20030516-utc/r
Created /export/20030516-utc/src/tools/obj/tools.NetBSD-1.6R-powerpc/bin/nbmake
Updated /export/20030516-utc/src/tools/obj/tools.NetBSD-1.6R-powerpc/bin/nbmake-macppc
Successful make release
build.sh ended: Thu May 29 14:32:28 UTC 2003
===> .
上の時間はとても短いが、これは make build が終った後からの時間。 一ヶ所間違った変更をしていたので、途中で止ってしまい、 setenv BUILD_DONE yes してから続けた。

#4 [bind] view と secondary

僕も、いまこの問題に当っている。
一番安易には、(本末転倒の気もするが) 内部用 secondary と外部用 secondary を分ければ済む話 ?

rsync とかで写した方がいいのでは ? という気がして来た。


#5 [iPod] iPod utility

こんなのがある。
Nota: you need a Windows iPod
何だこれ。


2003年05月28日(水) 旧暦 [n年日記]

#1 [Unix] tcsh dextract


bash で使えるか という話でなくて、それ自体、知らなかったというか、使ったことがない。 でも使うと便利そう。
そもそも cp ~1/file . なんて使い方もやったことがない。

#2 [機械] 一台 20030503ts に Upgrade (quick)

netbsd-GENERIC_MD.gz を /20030507.gz としておいておいて、 それで 起動 - sysinst

dns:

やはり DNS はまだ見に行ってくれない気がする。

/etc:

実は etcupdate というのを使ったことがないけれど、 新規設置の時に /etc の写しをそっくり持っていて OS を新しくしたら、 update -j old -j real /etc のようなことが出来た方が親切な気がする..
  OS         現用
--------     -----
/etc-1.6R -> /etc
というか sysinst の Upgrade は古い方の etc で起動して、 新しいのは /etc.new で置いておいて、 起動した後で調整するということ。

telnet で入った時に、文字入力が遅い ?:

ftp が 800kbps しか出ない ?:

1.6Q gm0 の機械相手に:
ftp> get /netbsd /dev/null
...
4455628 bytes received in 00:40 (106.18 KB/s)

diff -ur /etc /etc.new をしていて何故か遅かった。:

それを止めたら

88Mbps:

ftp> get /netbsd /dev/null
...
4455628 bytes received in 00:00 (11.20 MB/s)

#3 [MTA] 再び qpopper (4.0.5)

mail server を再び移 そうとしている。sendmail/qpopper/procmail
===> Please note the following:

===========================================================================
$NetBSD: MESSAGE,v 1.2 2003/03/13 07:21:42 itojun Exp $

In order to enable POP service, you should add the following lines
to your /etc/inetd.conf:

    pop3 stream tcp  nowait root /usr/pkg/libexec/qpopper qpopper -s
    pop3 stream tcp6 nowait root /usr/pkg/libexec/qpopper qpopper -s

===========================================================================

===> Registering installation for qpopper-4.0.5
ttyp0:makoto@fml  10:20:06/030528(...mail/qpopper)> 


2003年05月27日(火) 旧暦 [n年日記]

#1 [MTA] 知合いからメール届かずと電話あり

ttyp4:root@ibook 12:52:50/030527(/home/makoto)# grep hogesys /var/log/maillog
May 27 04:01:33 ibook sendmail[18239]: h4QIr2GY018239: collect: premature EOM: Connection reset by mon.hogesys.co.jp
May 27 04:01:34 ibook sendmail[18239]: h4QIr2GY018239: SYSERR(root): collect: I/O error on connection from mon.hogesys.co.jp, from=<mymy@hogesys.co.jp>
May 27 04:01:34 ibook sendmail[18239]: h4QIr2GY018239: from=<mymy@hogesys.co.jp>, size=1005, class=0, nrcpts=1, proto=SMTP, daemon=MTA, relay=mon.hogesys.co.jp [hh.221.yy.xxx]
May 27 12:36:11 ibook sendmail[19417]: h4R3RdGY019417: collect: premature EOM: Connection reset by mon.hogesys.co.jp
May 27 12:36:11 ibook sendmail[19417]: h4R3RdGY019417: SYSERR(root): collect: I/O error on connection from mon.hogesys.co.jp, from=<mymy@hogesys.co.jp>
May 27 12:36:11 ibook sendmail[19417]: h4R3RdGY019417: from=<mymy@hogesys.co.jp>, size=1005, class=0, nrcpts=1, proto=SMTP, daemon=MTA, relay=mon.hogesys.co.jp [hh.221.yy.xxx]
まさか、こういうのではないと思うけれど..
If you are using a Cisco firewall product that supports their fixup protocol,
try disabling it for smtp. As far as I know, it only supports SMTP and you may be using ESMTP.
ipnat の rdr 機能を使っていると、この問題があるという気がする。
ipf: IP Filter: v3.4.29 (344)
Kernel: IP Filter: v3.4.29 ~ Running: yes
Log Flags: 0 = none set
Default: pass all, Logging: available
Active list: 0
ではなくて ipnat の方か。
@(#)ipnat.c 1.9 6/5/96 (C) 1993 Darren Reed
@(#)Id: ipnat.c,v 2.16.2.21 2002/06/06 10:49:19 darrenr Exp


2003年05月26日(月) 旧暦 [n年日記]

#1 [Web] hover

css に書く次のもの... 知らなかった。色が変る。
a:hover {background-color:#ffddee;color:#ff0000;}

#2 [NetBSD] pkgsrc/ap2-auth-postgresql

長い道のりだったが、Apache2 でやっと AuthPG が動いた。忘れないうちに覚えを書いておかなくては。
  1. pkgsrc/ap-auth-postgresql は AuthPG-1.2 で Apache2 には対応していない
  2. ( Mailing List に入って見たら)、 2002/04 頃に Apache2 対応の patch を流した人がいた。
  3. それを使って、見よう見真似で pkgsrc を作って見る
  4. .htaccess を用意する
  5. 実は(僕は) postgresql を tcp/ip で話せるように設定をしたことがなかった。
  6. それは postgresql.conf と pg_hba.conf
    に少しまじないを書けばいいのだが、 それをどこに置くのかの情報が(僕には)なかなか見つからなかった。 /usr/pkg/etc/postgresql /usr/pkg/share/postgresql ではなかった。(と思う)
    (もっとも postgresql.conf に
    tcpip_socket = on
    と書く方は、 起動時に -i を付ければ、なくても良かったのだけれど)
  7. ktrace
    cd /tmp
    ktrace /usr/pkg/bin/postmaster -D directory
    (/tmp)# kdump ktrace.out | grep postgresql.conf
    (/tmp)# kdump ktrace.out | grep pg_hba.conf
    
    のようにして、 ktrace/kdump を使って調べてしまった。
  8. 置くべき位置は、postmater 起動時に -D で指定した directory
    (それが current-directory になる)
  9. で、postgres 的にも www という user を作っておく必要がある
    createuser www
    (ま、これは apache2 が www で動いていれば当然のことである)
  10. grant もしておく必要がある
    (これがないと ERROR: hogehoge: permission denied )
    recent_status=# GRANT SELECT ON hogehoge TO PUBLIC;
    GRANT
    (など)

何故長いかって、AuthPG は ChiBUG で mira さんに教わって以来だし。
(でも実は AuthPG のどこが嬉しいかって、良く分っていない)。


2003年05月25日() 旧暦 [n年日記]

#1 [pkgsrc] apache2/php4/postgresql

作り直して見る 「devel/php4-session しただけでは --enable-session してくれなくて、 www/php4/Makefile.php を変更する必要がある」 気がする。


2003年05月24日() 旧暦 [n年日記]

#1 [Emacs] selecting deleted buffer

wl-2.10.0 で、例えば、namazu で検索して次のような時に、上から見ていって 8 のところで出た。
    1 N02/02(日)04:36 [ burt             ] ipnat.conf for darwin streaming serv
    2 N12/08(日)10:17 [ Masahiro Yamagis ] #08272 Re: question ipf             
   10  08/09(金)17:20 [ MOCHIDA Shuji    ] #07938 mssclamp on PPPoE network
    6  08/10(土)00:01 ┣[ FUKAUMI Naoki    ]                               
    4  08/11(日)23:54 ┃┣[ MOCHIDA Shuji    ]                             
    5  08/10(土)07:45 ┃┗[ 藤原 誠 / Makoto ]                             
    3  08/12(月)00:02 ┃ ┗[ MOCHIDA Shuji    ]                           
    8 N08/09(金)18:15 ┣[ Masaru OKI       ]                               
    7 N08/09(金)18:48 ┃┗[ MOCHIDA Shuji    ]               
    ....
これは、この状態で何度でも言うので、 (setq debug-on-error t) してからもう一度 8 を見ると、次のような字がある。
byte-code("......" [err wl-message-buffer-cache-delete signal] 3)
wl-message-buffer-display([elmo-nmz-folder [0 0 0 0 0 0 0] nmz "[mssclamp]" "[" "/home/makoto/.elmo/nmz/[mssclamp]"

wl-message.el:

158 (defun wl-message-buffer-cache-delete (&optional key)
159   "Delete the most recent cache entry"
160   (if key
161       (setq wl-message-buffer-cache
162             (delq (assoc key wl-message-buffer-cache)
163                   wl-message-buffer-cache))
164     (let ((buf (wl-message-buffer-cache-buffer-get
165                 (car wl-message-buffer-cache))))
166       (setq wl-message-buffer-cache
167             (nconc (cdr wl-message-buffer-cache)
168                    (list (wl-message-buffer-cache-entry-make nil buf)))))))
これは原因でなくて結果かなぁ。

wl-message-buffer-cache-size:

10

wl-message-buffer-cache:

((("[mssclamp]" 3 "<20020812.000222.125128237.mochid at netside.co.jp>") . #<buffer  *WL:Message*<5>>) 
 (("[mssclamp]" 5 "<yfmvg6jwt1u.wl at u.ki.nu>") . #<buffer  *WL:Message*<4>>) 
 (("[mssclamp]" 4 "<20020811.235400.65654786.mochid at netside.co.jp>") . #<buffer  *WL:Message*<3>>) 
 (("[mssclamp]" 6 "<874re4oz3n.wl at dns1.fukaumi.org>") . #<buffer  *WL:Message*<2>>) 
 (("[mssclamp]" 10 "<20020809.172010.65657653.mochid at netside.co.jp>") . #<buffer  *WL:Message*>) 
 (("[mssclamp]" 2 "<20021208100442.9858.NIGHT at pluto.dti.ne.jp>") . #<buffer  *WL:Message*<9>>) 
 (("[mssclamp]" 1 "<6F7FA4D8-361C-11D7-8F55-0030654D9470 at wanadoo.fr>") . #<buffer  *WL:Message*<8>>) 
 (("+ml/netbsd" 6560 "<20030518.182817.68562309.toshi at odd.minolta.co.jp>") . #<buffer  *WL:Message*<7>>) 
 (("+ml/netbsd" 6567 "<20030523.234324.87582895.taca at back-street.net>") . #<buffer  *WL:Message*<6>>) 
(nil . #<killed buffer>))
8 はまだ見ていないのに、あるはずと(いう人がいるので)見に行くと、「それはもう消したやつだよ」と言われている のかな。

そうではなくて、「buffer-cache が一杯になったので、 古いやつを消そうとした時に、以前に消したやつをまた消そうとしている」 かな

上のは単に結果のようだ。:

465 ;; Use message buffer cache.
466 (defun wl-message-buffer-display (folder number flag
....
489     (if (or force-reload read)
490         (condition-case err
...
501           (error
502            (wl-message-buffer-cache-delete)
503            (signal (car err) (cdr err))
504            nil))) ;; will not be used
505     (cons hit cache-used)))
そもそも 501 行目に来てしまうことが良くないらしい。

condition-case って何 ? appropos によれば:

Function: Regain control when an error is signaled.
Plist: lisp-indent-function byte-compile

wl-message.el:

 486       ;; delete tail and add new to the top.
 487       (setq hit (wl-message-buffer-cache-add (list fname number msg-id)))
 488       (setq read t))
ここで err になっている気がする。次のような message 行を付けておくと
486 ;; delete tail and add new to the top.
487 (message "hit false, cache adding fname(%s) number(%s) msg-id(%s)" fname number msg-id) ;; XXXX
488 (setq hit (wl-message-buffer-cache-add (list fname number msg-id)))
489 (message "hit(%s)" hit)
490 (setq read t))
hit false, cache adding fname([mssclamp]) number(8) msg-id(<20020809180839.26CA.OKI at netbsd.org>)
hit(#<killed buffer>)
と言っている。ここまで書けばよういちさんが何とかしてくれないかな。

(setq wl-message-buffer-cache nil):

すれば、Emacs を立上げ直さなくても大丈夫。

お手紙:

「Message-buffer を kill してしまったのに、それがあると思っているのがいけない」 ですよと御指名の方からお手紙をもらってしまいました。ありがたいことです。
  1. 自分で kill-buffer (C-x k) してしまった。(僕のことね)
  2. wl が何かの原因で kill-buffer してしまったのに、まだあると思って参照する
のどちらか。という切分けが必要とのこと。しかし上で書いた 8 のメールは とても古いやつで、きょうはまだ見ていないもの。 (検索で出て来た一覧だから、それは間違いない)。つまり二番目に当るはず。

「実は HTML を見るのに w3 を使っている」:

ことが重要かも知れないので、そうしています。とお知らせしておきます。

wl-message-buffer-cache:

の内容を見ると、確かにこの順に見た覚えがあるが、それの一つ前に見たメール を、自分で kill-buffer してしまったのかも知れない。 それは 上の 8 のものとは関係ない訳で、二つ上に書いた、 「とても古いやつで、きょうはまだ見ていないもの」というのは的外れだった。

つまり

「自分で kill-buffer (C-x k) してしまった。(僕のことね)」:

というのがありそうな話だ。 が、「(僕が)メールを見て、それを kill-buffer する」ことってあるかな、 と思ったりもする。記憶自信無。

(と書いたら)

「とりあえず trunk では対処しておきました。」という有難いお告げ:

試そうと思って、まずは wl-2.10.0 のままで:

再現法は簡単で、message buffer で C-x k して、あとは buffer cache がぐるぐる回ってバッファが存在しないとこに 足そうとするまで新しいメッセージを見るだけです。
しようとしたが、うまく行かない。(再現しない)。しかも 「message-buffer で C-x k 」するのって、ちょっと面倒で、 これはやっていないなぁ、という気が....

そうか、途中で cache の中を見ればいいのか。

どうも、単に kill しただけでは、上の方に書いた

 (nil . #<killed buffer>))
のように nill という字は入らない気がする。 (で、その時には再現しない ?)


2003年05月23日(金) 旧暦 [n年日記]

#1 [] PHPxPostgreSQL で作る 最強 Web システム

(技術評論社、石井達夫著) 2,880 円外税


2003年05月22日(木) 旧暦 [n年日記]

#1 [無題] 新宿南口方面

ソフマップ Mac/UMac:

1.2GHz 99,000 円。 (これは新品)高いなぁ。

14:00-18:00 まで某社で説明を聞く:



2003年05月20日(火) 旧暦 [n年日記]

#1 [NetBSD] apache-2.0.45 に VirtualHost を設定して見る

bind-9 の view を使って、特定の機械向けに 一時的な名前を設定出来るので、とても便利。 つまり、 仮の機械で VirtualHost を設定するのに、 本来の名前を全てそちらに向けたり出来る。
(でもそれだけなら、別に bind9 でなくても、一時的な DNS を動かせば いい気もするが)

#2 dvi2tty が pkgsrc にないので、/usr/local

に自分で patch/make (いや converters/dvi2tty はあるが、日本語対応していない気がしたので)。


2003年05月19日(月) 旧暦 [n年日記]

#1 [NetBSD] ffs_alloccg: map corrupted: start 2 len 510 offset 214 214

Calculated size of `ramdisk.fs.tmp': 2097152 bytes, 287 inodes
Extent size set to 4096
ramdisk.fs.tmp: 2.0MB (4096 sectors) block size 4096, fragment size 512
        using 1 cylinder groups of 2.00MB, 512 blks, 320 inodes.
super-block backups (for fsck -b #) at:
 32,nbmakefs: ffs_alloccg: map corrupted: start 2 len 510 offset 214 214

Populating `ramdisk.fs.tmp'
*** Error code 1

Stop.
nbmake: stopped in /export/20030516-utc/src/distrib/macppc/floppies/ramdisk
usr.sbin/makefs/ffs/ffs_alloc.c の次の行:
574     "ffs_alloccg: map corrupted: start %d len %d offset %d %ld",
575                                 ostart, olen,
576                                 ufs_rw32(cgp->cg_freeoff, needswap),
577                                 (long)cg_blksfree(cgp, needswap) - (long)cgp);
2048 を 4096 に変更して見たが、これは関係ない
ttyp0:root@serv 10:10:01/030519(...floppies/ramdisk)# (cd obj;/export/20030516-utc/src/tools/obj/tools.NetBSD-1.6R-powerpc/bin/nbmakefs -t ffs -B be -s 4096k -F work.spec -N /export/20030516-utc/src/etc -o bsize=4096,fsize=512 -o optimization=space,minfree=0 -f 15 ramdisk.fs.tmp work)
Calculated size of `ramdisk.fs.tmp': 4194304 bytes, 287 inodes
Extent size set to 4096
ramdisk.fs.tmp: 4.0MB (8192 sectors) block size 4096, fragment size 512
using 1 cylinder groups of 4.00MB, 1024 blks, 320 inodes.
super-block backups (for fsck -b #) at:
32,
Populating `ramdisk.fs.tmp'
nbmakefs: ffs_alloccg: map corrupted: start 2 len 1022 offset 214 214
ttyp0:root@serv 10:10:25/030519(...floppies/ramdisk)#
UPDATING/20030402:
The superblock layout for FFS was changed. If you have 1.6 fsck binaries, they will signal a fatal superblock mismatch
これと関係あるのかなぁ。

1.6O 辺りの makefs を持って来たら、正常終了風:

( cd obj ; /usr/sbin/makefs-1.6O -t ffs -B be -s 2048k -F work.spec -N /export/20030516-utc/src/etc -o bsize=4096,fsize=512 -f 20 ramdisk.fs.tmp work )
「makefs が変に新しくなってしまっている」ということ ?
static daddr_t ffs_mapsearch(struct fs *, struct cg *, daddr_t, int);  
       ↓
static int32_t ffs_mapsearch(struct fs *, struct cg *, daddr_t, int);
sys/arch/macppc/macppc/disksubr.c かなぁ。

#2 [mkid] Segmentation fault

ttyp0:makoto@serv  7:34:22/030519(...compat/darwin)> pwd
/export/20030516-utc/src/sys/compat/darwin
ttyp0:makoto@serv  7:34:25/030519(...compat/darwin)> mkid darwin_iof*c
Segmentation fault (core dumped)
ttyp0:makoto@serv  7:34:29/030519(...compat/darwin)> ls -l  darwin_iof*c
-rw-r--r--  1 makoto  wheel  26091 May 15 23:35 darwin_ioframebuffer.c

ttyp0:makoto@serv  7:36:56/030519(...compat/darwin)> gdb /usr/pkg/bin/mkid mkid.core
...
(gdb) where 
#0  0x72466c60 in ?? ()
とても長い行があるのが原因風。
/usr/include/stdio.h:#define BUFSIZ 1024 /* size of buffer used by setbuf */
これを使っているのかなぁ。BUFSIZ はかなりの数で参照されているので、直すのが面倒な気 がして、一時的に
/usr/include/stdio.h:#define BUFSIZ 65536 /* size of buffer used by setbuf */
にして作ったら、問題解決。
./libidu/idfile.c: static char file_name_buffer[BUFSIZ];
./libidu/fnprint.c: static char file_name_buffer[BUFSIZ];
./libidu/scanners.c: char id_0[BUFSIZ];
./libidu/scanners.c: char id_0[BUFSIZ];
./libidu/scanners.c: static char id_0[BUFSIZ];
./src/lid.c:char anchor_dir[BUFSIZ];
./src/lid.c: char line[BUFSIZ];
./src/lid.c: char buf[BUFSIZ];
./src/lid.c: char regexp_buf[BUFSIZ];
./src/lid.c: char ed_arg_buffer[BUFSIZ];
./src/lid.c: char pattern[BUFSIZ];
./src/lid.c: char buf[BUFSIZ];
./src/lid.c: static char pat_buf[BUFSIZ];
./intl/localealias.c: char buf[BUFSIZ];
./intl/localealias.c: if (fgets (buf, BUFSIZ, fp) == NULL)
./intl/localealias.c: if (fgets (buf, BUFSIZ, fp) == NULL)
このうち変更が必要なのは
./src/lid.c: char line[BUFSIZ];
だけかも知れない。いづれにしても lid.c の中だけでいい気がする。 は関係あるかな。
id-utils-3.2d を使う のが普通らしい


2003年05月17日() 旧暦 [n年日記]

#1 [機械] cvsweb が 15 日間くらい止っていた

amd の map を一部変更していて、 cvsweb.cgi の中の
$config = $ENV{'CVSWEB_CONFIG'} || '/usr/pkg/etc/httpd/cvsweb.conf';
で指定されている中の、
25-%CVSROOT = (
26-#        'Development' => '/usr/local/src/cvsrep',
27:         'misc' => '/a/j/public-repository',
28:         'tracking'        => '/a/j/tracking',
29-#    'HTML-files' => '/tmp/upload'
30-         );
31
の先が有効になっていなかった。


2003年05月16日(金) 旧暦 [n年日記]

#1 [NetBSD] macppc -> /dev/battery

major = 29:

ttyp0:root@harry 8:32:59/030516(...macppc/dev)# grep battery ../conf/majors.macppc
device-major battery char 29 battery

mknod /dev/battery c 29 0:

ls -lag /dev/battery
crw-r--r-- 1 root wheel 29, 0 May 16 08:27 /dev/battery

changing CPU clock:

(benchmark 1000
(format-time-string "\
%a%A%b%B%c%C%d%D%e%h%H%I%j%k%l%m%M%n%p%r%R%s%S%t%T%U%w%W%x%X%y%Y%Z"))

0.27202606201171875

echo power1 > /dev/battery
0.25034594535827637
0.25053393840789795
0.25037896633148193

echo power0 > /dev/battery
0.21664094924926758
0.216575026512146
0.21674096584320068
0.21651005744934082
多分 667MHz と 800MHz
echo suspend > /dev/battery
すると止ってしまう。どうやって起きるのか ?

ac on, 96% of battery:

ttyp0:root@harry 8:31:42/030516(...macppc/dev)# cat /dev/battery0 1 96 0

UFS と UFS2:

inode UFS(128) UFS2(256) ?

echo power0 > /dev/battery:

time make
...
4023468 74184   238368  4336020 422994  netbsd
393.069u 53.654s 7:25.76 100.2% 0+0k 10+2801io 39pf+0w
ttyp4:makoto@harry  10:12:51/030516(...compile/INSECURE_ZS_ABORT)> 

echo power1 > /dev/battery:

make clean && make depend && time make
...
4023468 74184   238368  4336020 422994  netbsd
460.482u 59.373s 8:24.21 103.1% 0+0k 0+2892io 0pf+0w
ttyp4:makoto@harry  10:25:15/030516(...compile/INSECURE_ZS_ABORT)> 

StarMax + G3/240 (cpu0: L2 cache not enabled):

2120.849u 360.393s 41:58.59 98.5%       0+0k 32+7157io 309pf+4w
ttyp0:makoto@fml  20:12:34/030516(...compile/INSECURE)> 


2003年05月15日(木) 旧暦 [n年日記]

#1 [NetBSD] 1.6Q -> 1.6R

世の中は 1.6S -> 1.6T らしいが
ttyp1:makoto@harry 7:13:16/030515(...macppc/conf)> config INSECURE
../../../../dev/ata/files.ata:17: syntax error
../../../../dev/ata/files.ata:21: undefined attribute `ataraid'
../../../../arch/macppc/conf/files.macppc: device `ataraid' used but not defined
usr.sbin/config 入替

#2 [pkgsrc] fml

単なる覚え
# pkg_delete fml
pkg_delete: unable to completely remove directory '/usr/pkg/fml/sbin'
pkg_delete: unable to completely remove directory '/usr/pkg/fml/module/Jcode/Jcode'
pkg_delete: unable to completely remove directory '/usr/pkg/fml/module/Jcode'
pkg_delete: unable to completely remove directory '/usr/pkg/fml/module'
pkg_delete: unable to completely remove directory '/usr/pkg/fml/drafts/Japanese'
pkg_delete: unable to completely remove directory '/usr/pkg/fml/drafts'
pkg_delete: unable to completely remove directory '/usr/pkg/fml/.fml'
pkg_delete: unable to completely remove directory '/usr/pkg/fml'
pkg_delete: couldn't entirely delete package `fml-4.0.20030419'
(perhaps the packing list is incorrectly specified?)
===============================================================

  /var/spool/ml won't be removed by pkg_delete, to keep infor-
  mation at 'make update'.

  Please remove that directory if you really want to uninstall 
  the package by:
     rm -rf /var/spool/ml

  There may also be some fml entries in /etc/mail/aliase file.
  Deleting those lines may be necessary for completion.
===============================================================
# 
上とは関係ないが
  • 20030419 は既に無い
  • pre-install: という標的は(当然) make install するまでは 実行されない。
    (何が困るかと言って files/japanese.sed が当ったか make install する前には調べられない)

#3 [機械] Stealth serial G4 を使って見る

OF からの serial は 57600bps らしい

cannot mount root, error = 79:

> boot hd:,ofwboot.xcf 20030503_MD.gz で
bootpath: /pci@f2000000/pci-bridge@d/mac-io@7/ata-4@1f000/disk@0
...
ums0 at uhidev2: 3 buttons and Z dir.
wsmouse0 at ums0 (mux ignored)
boot device: wd0
root on wd0a dumps on wd0b
no file system for wd0 (dev 0xa00)
cannot mount root, error = 79
root device (default wd0a): md0a
use one of: wd0[a-p] cd0[a-p] gm0 fw0 ddb halt reboot
root device (default wd0a):
MD, つまり Memory disk になっていないような気がする。
その通りだった。しかし 途中から速度が、多分 38.4k に変化 するような..

media 100baset-fdx:

autoselect は効かなかった。

gzip: unknown option -- l:

これは前にも見た
     Status: Running
    Command: progress -zf /mnt2/man.tgz tar -xepf -

----------------------------------------------------------
gzip: unknown option -- l
gzip: can't gzopen /mnt2/man.tgz.gz: Read-only file system
  9208 KB    2.98 MB/s

/mnt/sbin/pdisk:

pdisk が GENERIC_MD.gz に入っている
# pdisk /dev/rwd0c
Edit /dev/rwd0c -
Command (? for help): P

Partition map (with 512 byte blocks) on '/dev/rwd0c'
 #:                type name               length   base      ( size )
 1: Apple_partition_map Apple                  63 @ 1        
 2:      Apple_Driver43*Macintosh              54 @ 64       
 3:      Apple_Driver43*Macintosh              74 @ 118      
 4:    Apple_Driver_ATA*Macintosh              54 @ 192      
 5:    Apple_Driver_ATA*Macintosh              74 @ 246      
 6:      Apple_FWDriver Macintosh             200 @ 320      
 7:  Apple_Driver_IOKit Macintosh             512 @ 520      
 8:       Apple_Patches Patch Partition       512 @ 1032     
 9:           Apple_HFS 名称未設定          65536 @ 1544      ( 32.0M)
10:           Apple_HFS 名称未設定 2      2048000 @ 67080     (1000.0M)
11:          Apple_Boot MOSX_OF3_Booter     16384 @ 2115080   (  8.0M)
12:        Apple_Loader SecondaryLoader      1024 @ 2131464  
13:           Apple_UFS Mac OS X         61422592 @ 2132488   ( 29.3G)
14:     Apple_UNIX_SVR2 A/UX Root         1024000 @ 63555080  (500.0M) S0 RUFS k0
15:     Apple_UNIX_SVR2 Swap              2048000 @ 64579080  (1000.0M) S1  SFS K0
16:     Apple_UNIX_SVR2 Usr file system   4096000 @ 66627080  (  2.0G) s3   FS   
17:     Apple_UNIX_SVR2 Unreserved 1     85577728 @ 70723080  ( 40.8G) s3   FS   
18:          Apple_Free                       679 @ 156300808

Device block size=512, Number of Blocks=156301487 (74.5G)
DeviceType=0x0, DeviceId=0x0
Drivers-
1:  23 @ 64, type=0x1
2:  36 @ 118, type=0xffff
3:  21 @ 192, type=0x701
4:  34 @ 246, type=0xf8ff

Command (? for help): 


2003年05月14日(水) 旧暦 [n年日記]

#1 [無題] 日赤健診・秋葉原・日赤・荻窪・幕張

日赤健診 前回 2002/10
秋葉原 Sony ICF-A55
日赤 眼鏡を忘れて取りに
荻窪 一枚書類が不足していて
幕張 ガーデン・ウォーク
BGM = KV.545

#2 [tcode] へ + 形 = 診 ?

へ + 形 = 診

#3 [日記]



2003年05月13日(火) 旧暦 [n年日記]

#1 [機械] 某なつみかんサーバの /home が

別のところを指していたので、5/1 .. 5/13 の間、 更新されていなかった。

#2 [pkgsrc] bootstrap-pkgsrc

実はきょうまで bootstrap-pkgsrc の場所を知らなかった。


2003年05月12日(月) 旧暦 [n年日記]

#1 [無題] 荻窪方面から菜の花プラザ



2003年05月11日() 旧暦 [n年日記]

#1 [無題] 15 人くらい集まる

  • 白木屋文書 台所初年目 友八の儀 (油井宏子さん)
  • 語り物文化の近代 「映画説明」レコードとその周辺(渡辺裕さん)

#2 PowerMac4400/200 +G3/240 到着



2003年05月10日() 旧暦 [n年日記]

#1 [無題] 君津出口 (4/29)

(館山自動車道) 4/29 日に開通した君津出口から初めて出た。 2.5km の延長のはずが、料金が同じ。少し得した気分。

#2 [機械] 機械をいくつか移動する

一階から機械室に移動したり、 一年間仮住いだった本箱の上から元の棚に戻す。


2003年05月09日(金) 旧暦 [n年日記]

#1 [Abuse] openfind

robot2.openfind.com.tw - - [09/May/2003:07:52:21 +0900] "GET /%7Emakoto/diary/?200212a&amp;to=200212050 HTTP/1.0" 200 18780 "-" "Openbot/3.0+(robot-response@openfind.com.tw;+http://www.openfind.com.tw/robot.html)"

#2 [機械] G4/400 AGP 多分 M7641

到着。 10.2 を入れるのに一時間半くらいかかる。IBM の 10G はだめ。


2003年05月08日(木) 旧暦 [n年日記]

#1 [NetBSD] panic: rt_llinfo(0xe02d8620) is not equal to ln(0xe00d8620)

1.6O + 1.6R (kernel+fsck_ffs) machine
panic: rt_llinfo(0xe02d8620) is not equal to ln(0xe00d8620)
Stopped in pid 7.1 (ioflush) at netbsd:cpu_Debugger+0x14: mtlr 0
db>
しまった。trace を入忘れ。

#2 [無題] 本厚木方面で FPGA 板と戯れ

昼御寿司、夜蟹づくし、をごちそうになる

#3 [tcode] 虚 + 我 = 戯

    ○                     虚 = {上, セ, 並}
・●・・      ・・・・
・・・・      ・・・・     ●…第1打鍵
・・・・      ・・・・     ○…第2打鍵
    ●                     我 = {戈, 禾}
・・・・      ・・・・
○・・・      ・・・・     ●…第1打鍵
・・・・      ・・・・     ○…第2打鍵
殆んど入力することがないなぁ。


2003年05月07日(水) 旧暦 [n年日記]

#1 [NetBSD] vnconfig

実は vnconfig を使ったことがなかった
mkisofs -r -J -l -o CDimage CD
vnconfig /dev/vnd0c CDimage
mount -t cd9660 -r /dev/vnd0c /mnt
ls /mnt
...
umount /mnt
vnconfig -u /dev/vnd0c

#2 [network] Yahoo BB

いつもお世話になっている人から電話で、
Yahoo BB だと年に 30k 円くらいで、今 安房ネット に払っているのが 50k /年 NTT さんに 3.5k x 12 で、合計 42k /年なので Yahoo BB の方が安いのでは ? というお話があるが、どう思うか

とのこと。50k が 30k になるだけだと思う。多分。

#3 [Emacs] benchmark.el

G4/400 0.27 (emcws):

(benchmark 1000
(format-time-string "\
%a%A%b%B%c%C%d%D%e%h%H%I%j%k%l%m%M%n%p%r%R%s%S%t%T%U%w%W%x%X%y%Y%Z"))

0.27530503273010254
0.2736409902572632
0.27215099334716797
0.2714120149612427
0.3528909683227539
0.28250598907470703

G4/667 (と信じているやつ) 0.18 (emcws):

(benchmark 1000
(format-time-string "\
%a%A%b%B%c%C%d%D%e%h%H%I%j%k%l%m%M%n%p%r%R%s%S%t%T%U%w%W%x%X%y%Y%Z"))

0.18636095523834229
0.18615007400512695
0.18498694896697998
0.1846550703048706
0.18544507026672363
0.1882469654083252

G4/800 0.05 (ただし emacs):

(benchmark 1000
(format-time-string "\
%a%A%b%B%c%C%d%D%e%h%H%I%j%k%l%m%M%n%p%r%R%s%S%t%T%U%w%W%x%X%y%Y%Z"))
0.05885791778564453
0.0817110538482666
0.05786001682281494
0.07932400703430176
0.057554006576538086

G4/800 0.15 (emcws 21.3):

(benchmark 1000
(format-time-string "\
%a%A%b%B%c%C%d%D%e%h%H%I%j%k%l%m%M%n%p%r%R%s%S%t%T%U%w%W%x%X%y%Y%Z"))
0.1584700345993042
0.1583549976348877
0.15877199172973633
0.15908801555633545
0.1632070541381836

#4 [Emacs] tcode

error: "ファイル /home/makoto/T-Code2/bushu.index が存在しません。"
これはどうい時に出るのかなぁ。
Symbol's function definition is void: tcode-set-key
site-lisp を全とっかえしたら直った。

#5 [lpr]

ttyp8:makoto@harry 22:20:55/030507(~)> lpr mozilla.ps
lpr: mozilla.ps: copy file is too large (check :mx:?)
lpr: connect: No such file or directory
jobs queued, but cannot start daemon.


2003年05月05日(月) 旧暦 [n年日記]

#1 [NetBSD] build.sh 20030502-ts

cc -O2 -pipe -Werror -Wall -Wpointer-arith -Wstrict-prototypes -Wmissing-prototypes -Wno-uninitialized -Wno-main -msoft-float -ffreestanding -I. -I/export/20
030502ts/src/sys/arch -I/export/20030502ts/src/sys -nostdinc -DDIAGNOSTIC -DMAXUSERS=32 -D_KERNEL -D_KERNEL_OPT -Dmacppc -c /export/20030502ts/src/sys/arch/ma
cppc/macppc/disksubr.c
/export/20030502ts/src/sys/arch/macppc/macppc/disksubr.c:509: conflicting types for `readdisklabel'
/export/20030502ts/src/sys/sys/disklabel.h:501: previous declaration of `readdisklabel'
*** Error code 1

Stop.
nbmake: stopped in /export/20030502ts/src/sys/arch/macppc/compile/GENERIC

src/sys/arch/macppc/macppc/disksubr.c:

507 const char *
508 readdisklabel(dev, strat, lp, osdep)
509         dev_t dev;
510         void (*strat)(struct buf *);
511         struct disklabel *lp;
512         struct cpu_disklabel *osdep;

src/sys/sys/disklabel.h:

500 char    *readdisklabel __P((dev_t, void (*)(struct buf *), struct disklabel *,
501             struct cpu_disklabel *));
const が付いただけかな。1.81 になれば付いている。
cvs update -A src/sys/sys/disklabel.h
(もしかして椿井さんが先に合わせた ?)

ZS_CONSOLE_ABORT が GENERIC に入っている:

:

cc -msoft-float -ffreestanding -O2 -Wreturn-type -Werror -Wall -Wno-main -Wno-format-zero-length -Wpointer-arith -Wmissing-prototypes -Wstrict-prototypes -Wno-sign-compare -Wno-uninitialized -Dmacppc -I. -I../../../../arch -I../../../.. -nostdinc -DLKM -DDIAGNOSTIC -DZS_CONSOLE_ABORT -DMAXUSERS=32 -D_KERNEL -D_KERNEL_OPT -c ../../../../dev/ccd.c
cc1: warnings being treated as errors
../../../../dev/ccd.c: In function `ccdgetdisklabel':
../../../../dev/ccd.c:1418: warning: assignment discards qualifiers from pointer target type
*** Error code 1
cvs update -D 20030504-UTC sys/dev/ccd.c
cvs update -D 20030504-UTC sys/dev/vnd.c

#2 [機械] 一台機械室に移動

3000/200 に入っている 1.6Q を物理的に移動

#3 [無題] 錦糸町からMOA+

都立現代美術館 にて サム・フランシス展
都バスにて白河二丁目から秋葉原まで
結局何も買わずにそのまま帰る。


2003年05月04日() 旧暦 [n年日記]

#1 [無題] 本: GIMP エッセンシャルガイド

八重洲ブックセンター → 京橋 → 銀座 → 歌舞伎座の夜の部 と全て徒歩。 團菊祭 (「髪結新三」等)


2003年05月02日(金) 旧暦 [n年日記]

#1 [Network] Router 入替

一年近く使って来た (入口の) Router を入替えた。
StarMax 3000/180 + G3/200, NetBSD/macppc 1.6Beta 
                          ↓
StarMax 3000/180           NetBSD/macppc 1.6O + 1.6Q
(StarMax 3000/180 = PPC603e/180MHz)
何だか二倍くらい速くなった気がする (多分気のせい)。
G4/400 に ftp して get /netbsd /dev/null すると、4.51 MB/s = 36Mbps だから、 やはり気のせいか。 ftp.dti.ad.jp から get ls-lR.gz /dev/null すると 1.17 MB/s = 9.6Mbps (これは前と同じ)

起動してから交換は可 ?:

起動しておいてつなぎ変えてもいいのか、つないでおいて 起動する必要があるのか良く分っていない。


2003年05月01日(木) 旧暦 [n年日記]

#1 [Network] Cisco 2511

某事情により 2511 を入手。

Cisco reverse telnet:

11.0 では必要なかった "transport input telnet" が 11.2 では必要になっている。 Google がなかったら、これはただの箱 ?
どうも、この reverse telnet というのは モデムの設定や、障害切分用に用意 されているようだが、僕にとっては計算機管理用の便利な道具。



最近の日記
2024年05月08日
comparison on ./buildsh tools
2024年05月06日
py-setuptools (python 3.11.9)
make release took 1 hours and 10 min
qemu invocation for 10.99.10
2024年05月05日
Windows 10 version
serial connection
bc bench
2024年05月04日
Trial on 10.99.10
another version (later trial) to succeed
2024年04月29日
dkim
以上、27 日分です。
タイトル一覧
カテゴリ分類
Powered by hns-2.19.9, HyperNikkiSystem Project

Count.cgi (since 2000/02/05)