Re: How to run Microsoft Internet Explorer on NetBSD?

[coypu]

If you're content with VirtualBox, Microsoft themselves provide images
of various versions of Windows with various versions of IE already
installed.

https://dev.windows.com/en-us/microsoft-edge/tools/vms/windows/

Re: How to run Microsoft Internet Explorer on NetBSD?

[Eric Haszlakiewicz]

On January 29, 2016 5:31:17 AM EST, Jose Luis Rodriguez Garcia 
 wrote:
>I need to use Internet Explorer for access a web from my company:
>Siebel software.
>
>It uses Active X and it only works with Internet Explorer. (They
>haven't activated the mode of Siebel Software for non Microsoft
>browsers).
>
>.It is a application that I must use, very few times.
>
>It is the option that I have used (tested some of them). What option
>do you suggest?
>
>2- I can access with Citrix, but the citrix client from pkgsrc is very
>old ,and it has problems with certificates of our Citrix Server. I
>have tried to add the certificates without success. I don't know if I
>am doing the right thing, or the citrix client has some problem for
>understand the new certificates.

If you're running this from inside your company's network, Citrix is probably 
your best bet for minimizing friction with your IT/security department, and 
making things easy for yourself.  If you can get it the Citrix client working 
reliably then you can let someone else worry about Windows licensing, making 
sure the right version of IE is installed, security patches are installed, 
etc...

Eric

Re: How to run Microsoft Internet Explorer on NetBSD?

[Marina Brown]

On 01/29/2016 05:31 AM, Jose Luis Rodriguez Garcia wrote:
> I need to use Internet Explorer for access a web from my company:
> Siebel software.
> 
> It uses Active X and it only works with Internet Explorer. (They
> haven't activated the mode of Siebel Software for non Microsoft
> browsers).
> 
> .It is a application that I must use, very few times.
> 
> It is the option that I have used (tested some of them). What option
> do you suggest?
> 
> 1- Wine (I have tried it years ago with bad results with Explorer.
> 2- I can access with Citrix, but the citrix client from pkgsrc is very
> old ,and it has problems with certificates of our Citrix Server. I
> have tried to add the certificates without success. I don't know if I
> am doing the right thing, or the citrix client has some problem for
> understand the new certificates.
> 3- VMware.
> 4- Other emulator?
> 

I use Qemu for virtualization. Not much different that VMware but it is
open source if that matters in this instance. There is virtualbox that
you can use for virtualization too. If i were you i would keep a small
windows virt for use with your company - i used to do just the same.
They have a portal that requires windows. I think you can even get
windows 10 for free under some circumstances. Not sure about this i had
a windows license owned by my company.

--- Marina Brown

Re: How to run Microsoft Internet Explorer on NetBSD?

[Mayuresh]

On Fri, Jan 29, 2016 at 11:31:17AM +0100, Jose Luis Rodriguez Garcia wrote:
> I need to use Internet Explorer for access a web from my company:

www/ies4linux? [ I haven't used of late. ]

Mayuresh.

How to run Microsoft Internet Explorer on NetBSD?

[Jose Luis Rodriguez Garcia]

I need to use Internet Explorer for access a web from my company:
Siebel software.

It uses Active X and it only works with Internet Explorer. (They
haven't activated the mode of Siebel Software for non Microsoft
browsers).

.It is a application that I must use, very few times.

It is the option that I have used (tested some of them). What option
do you suggest?

1- Wine (I have tried it years ago with bad results with Explorer.
2- I can access with Citrix, but the citrix client from pkgsrc is very
old ,and it has problems with certificates of our Citrix Server. I
have tried to add the certificates without success. I don't know if I
am doing the right thing, or the citrix client has some problem for
understand the new certificates.
3- VMware.
4- Other emulator?

Re: How to run Microsoft Internet Explorer on NetBSD?

[Swift Griggs]

On Fri, 29 Jan 2016, Mayuresh wrote:

This may be an OT.


Nah, it's "user" related. Spot on the topic. :-)

I don't think using i386 is a bad idea either, unless one has some 
specific reason to use amd64.


I mostly agree with this sentiment.

I recently switched to amd64. I gained nothing, at least for my purpose, 
but lost wine.


I may be completely misremembering, or perhaps these reasons are merely 
superstitions, but I had a couple of reasons for opting for AMD64. Folks, 
please disabuse me of these notions if they are wrong.


* At some point I thought that PAE wasn't working properly on i386. My
  workstation machine has 24G of RAM (yeah yeah, but it was on sale). I
  have a foggy recollection of i386 only letting me access the first 4GB.

* I get slightly better performance on system benchmarks (again mostly the
  memory benchmarks) with AMD64. Interestingly, lately with NetBSD 7.0
  some benchmarks are beating Linux across the board on the same hardware
  (I use removable drives on the same workstation). I was very pleasantly
  surprised at that.

However,

* It seems like at some point flash was working on i386 but not on AMD64,
  but that may have been a long time ago.

* As you point out wine seems perpetually broken on AMD64. I'm not
  sure why, since it works on Linux x86_64. However, I'm not trying to
  make light of anyone's efforts. Everything takes work.

* Also some time in the past I had major USB bus error problems with AMD64
  but I'd try i386 and the system behaved fine.

All in all, Mayuresh, while I can see some reasons to run AMD64, I'm 
really on the same wavelength as you are - what's the point of hassling 
with AMD64 anyway ? Perhaps someone will point out the "killer feature" 
that we missed. :-)


-Swift

Re: How to run Microsoft Internet Explorer on NetBSD?

[Swift Griggs]

On Fri, 29 Jan 2016, Mayuresh wrote:

On Fri, Jan 29, 2016 at 11:31:17AM +0100, Jose Luis Rodriguez Garcia wrote:

I need to use Internet Explorer for access a web from my company:

www/ies4linux? [ I haven't used of late. ]


That is a useful package, but it won't work if you are on NetBSD 7.0 AMD64 
since Wine is -currently broken for AMD64, IIRC. Last time I checked was 
about a week ago.


Thanks,
  Swift

Re: How to run Microsoft Internet Explorer on NetBSD?

[Mayuresh]

On Fri, Jan 29, 2016 at 09:13:29AM -0700, Swift Griggs wrote:
> On Fri, 29 Jan 2016, Mayuresh wrote:
> >On Fri, Jan 29, 2016 at 11:31:17AM +0100, Jose Luis Rodriguez Garcia wrote:
> >>I need to use Internet Explorer for access a web from my company:
> >www/ies4linux? [ I haven't used of late. ]
> 
> That is a useful package, but it won't work if you are on NetBSD 7.0 AMD64
> since Wine is -currently broken for AMD64, IIRC. Last time I checked was
> about a week ago.

This may be an OT.

I don't think using i386 is a bad idea either, unless one has some
specific reason to use amd64.

I was a long term user of i386 and just to find out whether it makes any
difference (since I could not find any reasons why people use amd64,
except where they know they need a 64 bit system) I recently switched to
amd64. I gained nothing, at least for my purpose, but lost wine.

Mayuresh.

Re: gpt/fdisk and harddrive identified with 2 cylinders?

[Robert Elz]

darren...@yahoo.com said:
  | wd0: 
  | wd0: drive supports 16-sector PIO transfers, LBA48 addressing
  | wd0: 2794 GB, 5814021 cyl, 16 head, 63 sec, 512 bytes/sect x5860533168 
sectors 

That is too big for 32 bit labels, which means that you cannot use either 
fdisk or disklabel to manipulate it (and get access to all of its space).

It also most likely means that you cannot use sysinst (or not the regular
way) as sysinst's GPT support is poor.

Completely destroy the label sectors of the disk (with dd from /dev/zero,
which is in no way a "low level format", but never mind) and re-init it
using "gpt create", and avoid using fdisk or disklabel at all (if you
really need to, fdisk to just display the protective MBR is safe, but
nothing which edits the MBR)

Don't use disklabel, or mbrlabel, or any variant of any of those.

kre

gpt/fdisk and harddrive identified with 2 cylinders?

[Darren]

NetBSD armv7 7.0 NetBSD 7.0 (CUBIETRUCK.201509250726Z) evbarm

wd0 at atabus0 drive 0
wd0: 
wd0: drive supports 16-sector PIO transfers, LBA48 addressing
wd0: 2794 GB, 5814021 cyl, 16 head, 63 sec, 512 bytes/sect x 5860533168 sectors
wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
wd0(ahcisata0:0:0): using PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133) 
(using DMA)

armv7# fdisk /dev/wd0
Disk: /dev/wd0
NetBSD disklabel disk geometry:
cylinders: 2, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
total sectors: 32640, bytes/sector: 512

BIOS disk geometry:
cylinders: 2, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
total sectors: 32640

Partitions aligned to 16065 sector boundaries, offset 63

Partition table:
0: GPT Protective MBR (sysid 238)
    start 1, size 32639 (16 MB, Cyls 0/0/2-2/8/6)
    PBR is not bootable: Bad magic number (0x)
1: 
2: 
3: 
No active partition.
Drive serial number: 0 (0x)
armv7# 

armv7# gpt show wd0
   start    size  index  contents
   0   1 PMBR
   1   1 Pri GPT header
   2  32 Pri GPT table
  34  362513 
  362547  32 Sec GPT table
  362579   1 Sec GPT header
armv7# 

I've tried manually setting the cylinders to no avail.  I had recently done a 
lowlevel format of the drive with dd in=/dev/zero of=/dev/wd0.  Before that it 
was working great.  

Re: gpt/fdisk and harddrive identified with 2 cylinders?

[Michael van Elst]

darren...@yahoo.com (Darren) writes:

>wd0 at atabus0 drive 0
>wd0: 
>wd0: drive supports 16-sector PIO transfers, LBA48 addressing
>wd0: 2794 GB, 5814021 cyl, 16 head, 63 sec, 512 bytes/sect x 5860533168 sec=
>tors
>wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
>wd0(ahcisata0:0:0): using PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/1=
>33) (using DMA)

>armv7# fdisk /dev/wd0
>Disk: /dev/wd0
>NetBSD disklabel disk geometry:
>cylinders: 2, heads: 255, sectors/track: 63 (16065 sectors/cylinder)
>total sectors: 32640, bytes/sector: 512


Somewhere you have a confused label on that disk.
-- 
-- 
Michael van Elst
Internet: mlel...@serpens.de
"A potential Snark may lurk in every tree."

Re: How to run Microsoft Internet Explorer on NetBSD?

[Mayuresh]

On Fri, Jan 29, 2016 at 09:58:52AM -0700, Swift Griggs wrote:
> * I get slightly better performance on system benchmarks (again mostly the
>   memory benchmarks) with AMD64. Interestingly, lately with NetBSD 7.0
>   some benchmarks are beating Linux across the board on the same hardware
>   (I use removable drives on the same workstation). I was very pleasantly
>   surprised at that.

That deserves a great visibility. Could you share the benchmarks?

[ I use both Linux (at work) and NetBSD (at home). I find it difficult to
reason to myself, why don't I simply use Linux everywhere as it is faster,
has greater application availability etc. My instincts always tell me to
continue with NetBSD, but that's not `reasoning' really... ]

Mayuresh.

Re: Speaking of PAE

[John Nemeth]

On Jan 29, 10:23pm, Martin Husemann wrote:
} On Fri, Jan 29, 2016 at 02:09:17PM -0700, Swift Griggs wrote:
} > 
} > Can any of the kernel gods comment on if PAE is still considered unstable? 
} > It's hard to tell if it's disabled by default because it conflicts with 
} > the Xen memory model, or if it's because it causes stability issues.

 PAE is required for Xen, at least for dom0.  BTW, Xen questions
are best addresses to port-xen.

} Not unstable (AFAICT), but it comes with a performance penalty.
} 
} In some environments it makes sense (e.g. a i386 xen domU on a 64bit
} setup and little virtual memory allocated - I'm running one of those
} myself), but in most cases a straight i386 kernel (for max 4GB
} installed) or a straight amd64 kernel otherwise are preferable.
} 
} But if you go with a amd64 kernel, you still can install i386 userland
} and have all the same benefits you get with i386-pae. Modulo bugs in the
} compat_netbsd32 emulation the userland side of things should be the same.

 Also, differences in device numbers.  Be sure to use an amd64
version of MAKEDEV.

} There are a few minor issues (e.g. i386 savecore can not deal with a amd64
} kernel crash dump). We use this type of setup as default on mips64 and
} are trying to minimize the fallout.
} 
}-- End of excerpt from Martin Husemann

Re: How to run Microsoft Internet Explorer on NetBSD?

[Hal Murray]

swiftgri...@gmail.com said:
> All in all, Mayuresh, while I can see some reasons to run AMD64, I'm  really
> on the same wavelength as you are - what's the point of hassling  with AMD64
> anyway ? Perhaps someone will point out the "killer feature"  that we
> missed. :-) 

The thing you get with 64 bits is pointers that work in more than 4 gigabytes 
of memory.  That includes virtual memory as well as physical.  There are lots 
of cases where that is critical.  The usual one is databases.

On the other hand, all your pointers take twice as much memory.  That may 
have significant impacts on your cache footprint.

There are some problems where you need 64 bit integers.  You can get those on 
a 32 bit system with long long.  That may be a pain if you are trying to 
compile software that somebody else wrote.


swiftgri...@gmail.com said:
> * At some point I thought that PAE wasn't working properly on i386. My
>workstation machine has 24G of RAM (yeah yeah, but it was on sale). I
>have a foggy recollection of i386 only letting me access the first 4GB. 

I have amd64 7.0 running on a system with 4 GB but /proc/meminfo says it's 
only using 3.  Is there anything I have to do to get it to use the other GB?


-- 
These are my opinions.  I hate spam.

Re: How to run Microsoft Internet Explorer on NetBSD?

[Swift Griggs]

On Fri, 29 Jan 2016, Hal Murray wrote:
The thing you get with 64 bits is pointers that work in more than 4 
gigabytes of memory.


Yes, of course, but with PAE that shouldn't be a problem. Of course, it 
appears that PAE is still considered experimental in NetBSD, since it's 
not enabled by default.


There are some problems where you need 64 bit integers.  You can get 
those on a 32 bit system with long long.


If you are lucky... Without the new-school integer types in in stdint.h, 
those canonical declarations like "long" or even "long long" can be fuzzy.


That may be a pain if you are trying to compile software that somebody 
else wrote.


Agreed. That's why I wish things like uint16_t ,uint32_t, and uint64_t are 
so nice. However, I think they came with C99 or something like that. So, a 
lot of "legacy" code and code written by people who don't know about the 
new types (or don't care) still has older integer types. Programmers 
usually like determinism. I'm not sure why things went so far off the 
rails with integer sizes et al. Probably has to something to do with 
having so many compilers and CPU types in the 80's and 90's.


I have amd64 7.0 running on a system with 4 GB but /proc/meminfo says 
it's only using 3.  Is there anything I have to do to get it to use the 
other GB?


I've noticed the same thing. I always just figured it was the kernel 
reserving buffers or something. I guess if you switched on the 
experimental PAE code, it you might see a lot more of your RAM. Scope out 
this blog:


https://blog.netbsd.org/tnf/entry/pae_support_for_native_i386

I learned a thing or two when I read it just now and it speaks directly to 
this topic.


-Swift

Re: Speaking of PAE

[Martin Husemann]

On Fri, Jan 29, 2016 at 02:09:17PM -0700, Swift Griggs wrote:
> 
> Can any of the kernel gods comment on if PAE is still considered unstable? 
> It's hard to tell if it's disabled by default because it conflicts with 
> the Xen memory model, or if it's because it causes stability issues.

Not unstable (AFAICT), but it comes with a performance penalty.

In some environments it makes sense (e.g. a i386 xen domU on a 64bit
setup and little virtual memory allocated - I'm running one of those
myself), but in most cases a straight i386 kernel (for max 4GB
installed) or a straight amd64 kernel otherwise are preferable.

But if you go with a amd64 kernel, you still can install i386 userland
and have all the same benefits you get with i386-pae. Modulo bugs in the
compat_netbsd32 emulation the userland side of things should be the same.

There are a few minor issues (e.g. i386 savecore can not deal with a amd64
kernel crash dump). We use this type of setup as default on mips64 and
are trying to minimize the fallout.

Martin