lookup_nisplus module
Hello -
I'm new to this list so please forgive me if I do anything out of the norm. For the last year and a half I've been applying a fix the lookup_nisplus.so source to allow for non-standard nisplus maps. The map extension org_dir is hardcoded in the current and previous source as to always assume .org_dir to any preceding map name like passwd, hosts, etc. It occurred to me this morning that maybe I could get this changed in the native autofs package so that I don't have to keep patching rpms for every release of RHEL we use at work. Anyhow, I contacted Redhat and they stated this would be the proper place to suggest a patch. Below you'll find the patch to C source based on autofs 4.1.3. I know nisplus is probably rarely used on Linux in most environments, but unfortunately I'm stuck with it in mine. Please let me know if there is any way to have this patch verified, and possibly included in future releases.
regards,
Ryan Pisani
#autofs-nisplus.patch
--- autofs-4.1.3.orig/modules/lookup_nisplus.c 2004-01-29 10:01:22.000000000 -0
600
+++ autofs-4.1.3/modules/lookup_nisplus.c 2005-06-16
13:48:26.000000000 -0
500
@@ -74,14 +74,14 @@
debug(MODPREFIX "looking up %s", name);
- sprintf(tablename, "[key=%s],%s.org_dir.%s", name, ctxt->mapname,
+ sprintf(tablename, "[key=%s],%s.%s", name, ctxt->mapname,
ctxt->domainname);
result = nis_list(tablename, FOLLOW_PATH | FOLLOW_LINKS, NULL, NULL);
if (result->status != NIS_SUCCESS && result->status !=
NIS_S_SUCCESS) {
/* Try to get the "*" entry if there is one - note that we
*don'
t*
modify "name" so & -> the name we used, not "*" */- sprintf(tablename, "[key=*],%s.org_dir.%s", ctxt->mapname,
+ sprintf(tablename, "[key=*],%s.%s", ctxt->mapname,
ctxt->domainname);
result = nis_list(tablename, FOLLOW_PATH | FOLLOW_LINKS,
NULL, N
ULL);
}###END PATCH
Ian Kent
maintainerQuoted 6 lines · Ryan Pisani
Hello -
I'm new to this list so please forgive me if I do anything out of the norm. For the last year and a half I've been applying a fix the lookup_nisplus.so source to allow for non-standard nisplus maps. The map extensionorg_diris hardcoded in the current and previous source as to always assume .org_dir to any preceding map name like passwd, hosts, etc.
Quoted 9 lines
It occurred to me this morning that maybe I could get this changed in the native autofs package so that I don't have to keep patching rpms for every release of RHEL we use at work. Anyhow, I contacted Redhat and they stated this would be the proper place to suggest a patch. Below you'll find the patch to C source based on autofs 4.1.3. I know nisplus is probably rarely used on Linux in most environments, but unfortunately I'm stuck with it in mine. Please let me know if there is any way to have this patch verified, and possibly included in future releases.
Hi Ryan,
Thanks for your patch.
I have to admit my maintenance of the nisplus code has been poor. I've received a few patches over the last couple of years and haven't been able to merge any.
Priorities, not knowing nisplus myself and the complete lack of the ability to test has been the main problem. At one point I tried to work with someone on this but it rapidly became too hard.
atm the priority is getting the basic functionality expected of an automounter sorted out. Linux autofs just hasn't got it and I've been responsible for it for quite a while now so it's way past time to get it done.
So all I can say is I've filed it and hope to get to it.
Sorry, for now at least.
Ian
Quoted 44 lines
regards,
Ryan Pisani
#autofs-nisplus.patch
---autofs-4.1.3.orig/modules/lookup_nisplus.c2004-01-29 10:01:22.000000000 -0
600
+++autofs-4.1.3/modules/lookup_nisplus.c2005-06-16
13:48:26.000000000 -0
500(quoted diff) +1
-1@@ -74,14 +74,14 @@ debug(MODPREFIX "looking up %s", name); - sprintf(tablename, "[key=%s],%s.org_dir.%s", name, ctxt->mapname, + sprintf(tablename, "[key=%s],%s.%s", name, ctxt->mapname, ctxt->domainname); result = nis_list(tablename, FOLLOW_PATH | FOLLOW_LINKS, NULL, NULL); if (result->status != NIS_SUCCESS && result->status != NIS_S_SUCCESS) { /* Try to get the "*" entry if there is one - note that we *don't*
modify "name" so & -> the name we used, not "*" */(quoted diff) +1
-1- sprintf(tablename, "[key=*],%s.org_dir.%s", ctxt->mapname, + sprintf(tablename, "[key=*],%s.%s", ctxt->mapname, ctxt->domainname); result = nis_list(tablename, FOLLOW_PATH | FOLLOW_LINKS,NULL, N
ULL); }###END PATCH
_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs
Thanks Ian. Hopefully it can be used down the road. If you ever do need a tester, I'm always available, and I have access to multiple nisplus domains that differ from the primary one. I know nisplus is not a hot item in the world of linux and autofs. Thanks again.
Ryan
Quoted 98 lines · Ryan Pisani
Quoted 14 lines
Hello -
I'm new to this list so please forgive me if I do anything out of the norm. For the last year and a half I've been applying a fix the lookup_nisplus.so source to allow for non-standard nisplus maps. The map extensionorg_diris hardcoded in the current and previous source as to always assume .org_dir to any preceding map name like passwd, hosts, etc.
It occurred to me this morning that maybe I could get this changed in the
native autofs package so that I don't have to keep patching rpms for every
release of RHEL we use at work. Anyhow, I contacted Redhat and they statedQuoted 9 lines
this would be the proper place to suggest a patch. Below you'll find the patch to C source based on autofs 4.1.3. I know nisplus is probably rarely
used on Linux in most environments, but unfortunately I'm stuck with it in
mine. Please let me know if there is any way to have this patch verified,
and possibly included in future releases.Hi Ryan,
Thanks for your patch.
I have to admit my maintenance of the nisplus code has been poor. I've received a few patches over the last couple of years and haven't been able to merge any.
Priorities, not knowing nisplus myself and the complete lack of the ability to test has been the main problem. At one point I tried to work with someone on this but it rapidly became too hard.
atm the priority is getting the basic functionality expected of an automounter sorted out. Linux autofs just hasn't got it and I've been responsible for it for quite a while now so it's way past time to get it done.
So all I can say is I've filed it and hope to get to it.
Sorry, for now at least.
Ian
Quoted 48 lines
regards,
Ryan Pisani
#autofs-nisplus.patch
---autofs-4.1.3.orig/modules/lookup_nisplus.c2004-01-29 10:01:22.000000000 -0
600
+++autofs-4.1.3/modules/lookup_nisplus.c2005-06-16
13:48:26.000000000 -0
500(quoted diff) +1
-1@@ -74,14 +74,14 @@ debug(MODPREFIX "looking up %s", name); - sprintf(tablename, "[key=%s],%s.org_dir.%s", name, ctxt->mapname, + sprintf(tablename, "[key=%s],%s.%s", name, ctxt->mapname, ctxt->domainname); result = nis_list(tablename, FOLLOW_PATH | FOLLOW_LINKS, NULL, NULL); if (result->status != NIS_SUCCESS && result->status != NIS_S_SUCCESS) { /* Try to get the "*" entry if there is one - note thatwe
*don'
t*modify "name" so & -> the name we used, not "*" */- sprintf(tablename, "[key=*],%s.org_dir.%s",
ctxt->mapname,
+ sprintf(tablename, "[key=*],%s.%s",ctxt->mapname,
ctxt->domainname);
result = nis_list(tablename,FOLLOW_PATH|FOLLOW_LINKS,
NULL, NULL); }###END PATCH
_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs
Ian Kent
maintainerQuoted 4 lines · Ryan Pisani
Thanks Ian. Hopefully it can be used down the road. If you ever do need a tester, I'm always available, and I have access to multiple nisplus domains that differ from the primary one. I know nisplus is not a hot item in the world of linux and autofs. Thanks again.
Thanks Ryan. It's in my contrib directory.
What we really need is someone that wants to do a little nisplus development. But who has the time?
Ian
Quoted 110 lines
Ryan
Quoted 88 lines · Ryan Pisani
Quoted 14 lines
Hello -
I'm new to this list so please forgive me if I do anything out of the norm. For the last year and a half I've been applying a fix the lookup_nisplus.so source to allow for non-standard nisplus maps. The map extensionorg_diris hardcoded in the current and previous source as to always assume .org_dir to any preceding map name like passwd, hosts, etc.
It occurred to me this morning that maybe I could get this changed in the
native autofs package so that I don't have to keep patching rpms for every
release of RHEL we use at work. Anyhow, I contacted Redhat and they statedQuoted 9 lines
this would be the proper place to suggest a patch. Below you'll find the patch to C source based on autofs 4.1.3. I know nisplus is probably rarely
used on Linux in most environments, but unfortunately I'm stuck with it in
mine. Please let me know if there is any way to have this patch verified,
and possibly included in future releases.Hi Ryan,
Thanks for your patch.
I have to admit my maintenance of the nisplus code has been poor. I've received a few patches over the last couple of years and haven't been able to merge any.
Priorities, not knowing nisplus myself and the complete lack of the ability to test has been the main problem. At one point I tried to work with someone on this but it rapidly became too hard.
atm the priority is getting the basic functionality expected of an automounter sorted out. Linux autofs just hasn't got it and I've been responsible for it for quite a while now so it's way past time to get it done.
So all I can say is I've filed it and hope to get to it.
Sorry, for now at least.
Ian
Quoted 40 lines
regards,
Ryan Pisani
#autofs-nisplus.patch
---autofs-4.1.3.orig/modules/lookup_nisplus.c2004-01-29 10:01:22.000000000 -0
600
+++autofs-4.1.3/modules/lookup_nisplus.c2005-06-16
13:48:26.000000000 -0
500(quoted diff) +1
-1@@ -74,14 +74,14 @@ debug(MODPREFIX "looking up %s", name); - sprintf(tablename, "[key=%s],%s.org_dir.%s", name, ctxt->mapname, + sprintf(tablename, "[key=%s],%s.%s", name, ctxt->mapname, ctxt->domainname); result = nis_list(tablename, FOLLOW_PATH | FOLLOW_LINKS, NULL, NULL); if (result->status != NIS_SUCCESS && result->status != NIS_S_SUCCESS) { /* Try to get the "*" entry if there is one - note thatwe
*don'
t*modify "name" so & -> the name we used, not "*" */- sprintf(tablename, "[key=*],%s.org_dir.%s",
ctxt->mapname,
+ sprintf(tablename, "[key=*],%s.%s",ctxt->mapname,
ctxt->domainname);
result = nis_list(tablename,FOLLOW_PATH|FOLLOW_LINKS,
NULL, NULL); }Quoted 10 lines
Quoted 8 lines
###END PATCH
_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs_______________________________________________
autofs mailing list
[email protected]
http://linux.kernel.org/mailman/listinfo/autofs
Quoted 11 lines · Ian Kent
Quoted 4 lines · Ryan Pisani
Thanks Ian. Hopefully it can be used down the road. If you ever do need a tester, I'm always available, and I have access to multiple nisplus domains that differ from the primary one. I know nisplus is not a hot item in the world of linux and autofs. Thanks again.
Thanks Ryan. It's in my contrib directory.
What we really need is someone that wants to do a little nisplus development. But who has the time?
Ian,
In general nis+ development isn't actually needed. NIS+ isn't going anywhere and will probably be de-supported in the coming years. In the near future, however, when x86_64 platforms with virtualization becoming cheap and solaris x86 not being the red-headed stepchild at sun anymore, it becomes feasible to have a NIS+ infrastructure without needing a pile of sun boxes along with a linux client. The infrastructure becomes the requirement for a workstation cpu and a few dozen gigs of space on an ide drive.
Look for more action on this area next year when none of us have to stay in the office to fix NIS+ bugs (or, more accurately, none of us have to foobar our department NIS+ maps to test fixes).
-Peter
Signature · 3 lines
The 5 year plan: In five years we'll make up another plan. Or just re-use this one.