1 |
"J. Roeleveld" <joost@××××××××.org> writes: |
2 |
|
3 |
> On Tuesday, January 19, 2016 01:46:45 AM lee wrote: |
4 |
>> "J. Roeleveld" <joost@××××××××.org> writes: |
5 |
>> > On Monday, January 18, 2016 02:02:27 AM lee wrote: |
6 |
>> >> "J. Roeleveld" <joost@××××××××.org> writes: |
7 |
>> >> > On 17 January 2016 18:35:20 CET, Mick <michaelkintzios@×××××.com> |
8 |
>> >> > wrote: |
9 |
>> >> > |
10 |
>> >> > [...] |
11 |
>> >> > |
12 |
>> >> >>I use the icaclient provided by Citrix to access my virtual desktop at |
13 |
>> >> >>work, |
14 |
>> >> >>but have never tried to set up something similar at home. What |
15 |
>> >> >>opensource |
16 |
>> >> >>software would I need for this? Is there a wiki somewhere to follow? |
17 |
>> >> >> |
18 |
>> >> > I'd love to do this myself as well. |
19 |
>> >> > |
20 |
>> >> > Citrix sells the full package as 'XenDesktop'. To do it yourself you |
21 |
>> >> > need |
22 |
>> >> > a VMserver (Xen or similar) and a remote desktop tool that hooks into |
23 |
>> >> > the |
24 |
>> >> > VM display. (Spice or VNC) |
25 |
>> >> > |
26 |
>> >> > Then you need some way of authenticating users and providing access to |
27 |
>> >> > the |
28 |
>> >> > client software. [...] |
29 |
>> >> |
30 |
>> >> You would have a full VM for each user? |
31 |
>> > |
32 |
>> > Yes |
33 |
>> > |
34 |
>> >> That would be a huge waste of resources, |
35 |
>> > |
36 |
>> > Diskspace and CPU can easily be overcommitted. |
37 |
>> |
38 |
>> Overcommitting disk space sounds like a very bad idea. Overcommitting |
39 |
>> memory is not possible with xen. |
40 |
> |
41 |
> Overcommitting diskspace isn't such a bad idea, considering most installs |
42 |
> never utilize all the available diskspace. |
43 |
|
44 |
When they do not use it anyway, there is no reason to give it to them in |
45 |
the first place. And when they do use it, how do the VMs handle the |
46 |
problem that they have plenty disk space available, from their point of |
47 |
view, while the host which they don't know about doesn't allow them to |
48 |
use it? |
49 |
|
50 |
Besides, overcommitting disk space means to intentionally create a setup |
51 |
which involves that the host can run out of disk space easily. That is |
52 |
not something I would want to create for a host which is required to |
53 |
function reliably. |
54 |
|
55 |
And how much do you need to worry about the security of the VMs when you |
56 |
build in a way for the users to bring the whole machine, or at least |
57 |
random VMs, down by using the disk space which has been assigned to |
58 |
them? The users are somewhat likely to do that even unintentionally, |
59 |
the more the more you overcommit. |
60 |
|
61 |
> Overcommitting memory is, i think, on the roadmap for Xen. (Disclaimer: At |
62 |
> least, I seem to remember reading that somewhere) |
63 |
|
64 |
That would be a nice feature. |
65 |
|
66 |
>> >> plus having to take care of a lot of VMs, |
67 |
>> > |
68 |
>> > Automated. |
69 |
>> |
70 |
>> Like how? |
71 |
> |
72 |
> How do you manage a large amount of physical machines? |
73 |
> Just change physical to VMs and do it the same. |
74 |
> With VMs you have more options for automation. |
75 |
|
76 |
Individually, in lack of a better way. Per user when it comes to |
77 |
setting up their MUAs and the like, in lack of any better way. It |
78 |
doesn't make a difference if it's a VM or not, provided that you have |
79 |
remote access to the machine. |
80 |
|
81 |
When you one VM for many users, you install the MUA only once, and when |
82 |
you need to do updates, you do them only once. When you have many VMs, |
83 |
like one for each user, you have to install and update many times, once |
84 |
on each VM. |
85 |
|
86 |
>> >> plus having to buy a lot of Windoze licenses |
87 |
>> > |
88 |
>> > Volume licensing takes care of that. |
89 |
>> |
90 |
>> expensive |
91 |
> |
92 |
> Depends on the requirements. It's cheaper then a few hundred seperate windows |
93 |
> licenses. |
94 |
|
95 |
It's still more expensive than one, or than a handful, isn't it? |
96 |
|
97 |
>> >> and taking about a week to install the updates |
98 |
>> >> after installing a VM. |
99 |
>> > |
100 |
>> > Never heard of VM templates? |
101 |
>> |
102 |
>> It still takes a week to put the updates onto the template. |
103 |
> |
104 |
> Last time I had to fully reinstall a windows machine it took me a day to do |
105 |
> all the updates. Microsoft even has server software that will keep them |
106 |
> locally and push them to the clients. |
107 |
|
108 |
That would be useful to have. Where could I download that? |
109 |
|
110 |
Last time I installed a VM, it took a week until the updates where |
111 |
finally installed, and you have to check on it every now and then to |
112 |
find out if it's even doing anything at all. The time before, it wasn't |
113 |
a VM but a very slow machine, and that also took a week. You can have |
114 |
the fastest machine on the world and Windoze always manages to bring it |
115 |
down to a slowness we wouldn't have accepted even 20 years ago. |
116 |
|
117 |
>> >> Add to that that the xen host goes down at |
118 |
>> >> random time intervals (because the sending queue of the network card |
119 |
>> >> times out for reasons that cannot be determined) which can be as long as |
120 |
>> >> a day, a week or even up to three weeks, and you are likely to become a |
121 |
>> >> rather unhappy administrator. |
122 |
>> > |
123 |
>> > Sorry, but I consider that a bug in your hardware. If it's really that |
124 |
>> > unstable, replace it. |
125 |
>> > I've been running Xen enabled servers for nearly 15 years. Never had |
126 |
>> > issues |
127 |
>> > like that. If it were truly that unstable, it wouldn't be gaining |
128 |
>> > popularity. |
129 |
>> The hardware has already been replaced, and the problem persists. Other |
130 |
>> machines of identical hardware that don't run xen don't show any issues. |
131 |
> |
132 |
> I still say the hardware is buggy. With replacing, I meant replace it with |
133 |
> different hardware, not a different version of the same buggy stuff. |
134 |
|
135 |
The hardware is known to be 100% reliable by own experience for over a |
136 |
year, for all the machines. Only when xen is running, the problem shows |
137 |
up. |
138 |
|
139 |
Replacing the machine with another, identical one, allows to rule out |
140 |
that the particular machine which was replaced has an issue and was very |
141 |
easy to do, so that was a very reasonable second step after trying |
142 |
different network cards. Three different network cards, from three |
143 |
different manufactures, lead to the same error message. |
144 |
|
145 |
Googling the error message shows that quite a few ppl, with entirely |
146 |
different hardware, usually not running xen, have had the same message |
147 |
with very similar symptoms. |
148 |
|
149 |
This currently leaves these possibilities: |
150 |
|
151 |
|
152 |
1.) Xen doesn't work with this hardware. |
153 |
|
154 |
2.) The problem might somehow be caused by an SSD. |
155 |
|
156 |
3.) The error message is actually true and something yet unknown is |
157 |
going on on the network. |
158 |
|
159 |
4.) The problem may have been fixed a while ago in the kernel and has |
160 |
not been fixed in the xen kernel. |
161 |
|
162 |
5.) The gplpv drivers the VMs use cause the problem. |
163 |
|
164 |
6.) It's an issue with power management since the problem occurs when |
165 |
the machine and the VMs are not used/busy, at night. Disabling the |
166 |
power management for the network card has not made a difference, |
167 |
though. |
168 |
|
169 |
|
170 |
3.) is currently being worked on. It needs to be figured out and, if |
171 |
there's something weird going on, to be solved in any case. 6.) seems |
172 |
unlikely, 1.) and 2.) can be decided when the the hardware is replaced |
173 |
with something entirely different, which is the most painful and most |
174 |
time-consuming option. That would leave 4.) and 5.), and 3.) if 3.) |
175 |
cannot be resolved. |
176 |
|
177 |
It's easy to say that "the hardware is buggy". I'm not convinced that |
178 |
it is. In any case, you can always run into a situation in which xen |
179 |
doesn't work as well as you might wish or have experienced so far. |
180 |
|
181 |
>> >> Try kvm instead, and you'll find that |
182 |
>> >> it's impossible to migrate the VMs from xen to to kvm when you want to |
183 |
>> >> use virtio drivers because you can't install them on an existing Windoze |
184 |
>> >> VM. |
185 |
>> > |
186 |
>> > Not a problem with the virtualisation technology. It is an issue with |
187 |
>> > driver management inside MS Windows. |
188 |
>> > There are ways to migrate VMs succesfully, I just don't see the point in |
189 |
>> > wasting time for that. |
190 |
>> |
191 |
>> It's time consuming when you have to reinstall the VMs to migrate them |
192 |
>> to kvm. And when you don't have the installers of all the software |
193 |
>> that's on some of the VMs and can't get them, you either have to run |
194 |
>> them without virtio drivers or you can't migrate them. |
195 |
> |
196 |
> There are Howtos on the internet describing how to migrate VMs from 1 |
197 |
> technology to another. Shouldn't be too hard. |
198 |
|
199 |
I looked for them. Did you find one that tells you how to install |
200 |
the virtio drivers on an existing Windoze 7 VM and that actually works? |
201 |
It's already very difficult to get rid of gplpv drivers. |
202 |
|
203 |
> And keeping the installers at hand is, in my opinion, a requirement of sane |
204 |
> system management. |
205 |
> I have installers for all the versions of software I deal with. |
206 |
|
207 |
Indeed --- but some predecessor decided not to keep an installer which |
208 |
would be required and is now unavailable. So the only options are to |
209 |
leave the VM running under xen or to run it under KVM without virtio |
210 |
drivers. The latter is bad idea because the application the installer |
211 |
would be needed for already has severe performance problems built in, |
212 |
and making it worse isn't a good idea. |
213 |
|
214 |
>> > The biggest reason why I don't use KVM is the lack of full snapshot |
215 |
>> > functionality. Snapshotting disks is nice, but you end up with an unclean- |
216 |
>> > shutdown situation and anything that's not yet committed to disk is gone. |
217 |
>> |
218 |
>> I'm not sure what you mean. When you take a snapshot while the VM is not |
219 |
>> shut down, what difference does it make whether you use xen or kvm? |
220 |
> |
221 |
> A "snapshot" for KVM is ONLY the disks. |
222 |
> With Xen, VMWare and Virtualbox, I can also make a snapshot/copy of what's in |
223 |
> memory. It's that which makes the difference. |
224 |
|
225 |
Is that possible without freezing the VM while you make a snapshot of |
226 |
the memory? If not, how is it so much better than shutting the VM down? |
227 |
|
228 |
>> >> Then there's the question how well vnc or spice connections work over a |
229 |
>> >> VPN that goes over the internet. |
230 |
>> > |
231 |
>> > VNC works quite well, as long as you use a minimal desktop. (like |
232 |
>> > blackbox). Don't expect KDE or Gnome to be usable. |
233 |
>> > I haven't tried Spice yet, but I've read that it performs better. |
234 |
>> |
235 |
>> It's not like you had a choice when you have Windoze VMs. |
236 |
> |
237 |
> Windows has RDP, which is a lot better than VNC. Especially when dealing with |
238 |
> low-bandwidth connections. |
239 |
|
240 |
Wasn't RPD deprecated earlier in this discussion because it seemed to be |
241 |
not sufficiently secure? |
242 |
|
243 |
>> >> It's not like the employees could get |
244 |
>> >> reliable internet connections with sufficient bandwidth, not to mention |
245 |
>> >> that the company would have to get one in the first place, which isn't |
246 |
>> >> much easier to get, if any. |
247 |
>> > |
248 |
>> > That depends on where you are. |
249 |
>> |
250 |
>> In this country, you have to be really lucky to find a place where you |
251 |
>> can get a decent internet connection. |
252 |
> |
253 |
> Then in your country, working from home might not be the best option. |
254 |
|
255 |
That probably goes for most countries. |
256 |
|
257 |
>> > The company could host the servers in a decent datacentre, which should |
258 |
>> > take care of the bandwidth issues. |
259 |
>> |
260 |
>> And give all their data out of hands? And how much does that cost? |
261 |
> |
262 |
> I'm talking about putting your own hardware there, not letting the datacentre |
263 |
> company access to the servers. |
264 |
|
265 |
How could they reside in a datacenter without the ppl there having |
266 |
physical access to them? |
267 |
|
268 |
>> > For the employees, if they want to work from home, it's up to them to |
269 |
>> > ensure they have a reliable connection. |
270 |
>> |
271 |
>> It is as much problem of the company when they want the employees to |
272 |
>> work at home. And the employees don't have a choice, they can only get |
273 |
>> a connection they can get. |
274 |
> |
275 |
> If the company insists people work from home, they need to ensure the |
276 |
> employees have the option for a usable connection. Most companies I deal with |
277 |
> leave working from home as an option to the employees. |
278 |
|
279 |
Sometimes it's not an option, and there isn't anything a company could |
280 |
do to improve what internet connection an employee can get, unless |
281 |
they'd spend huge amounts of money to put cables or fiber glass into the |
282 |
ground, provided that they'd get the permissions for that. |
283 |
|
284 |
Sooner or later, it might become very difficult to find anyone who's |
285 |
still willing to spend all the time and money it takes to commute, or |
286 |
someone who can still afford it at all, and it might become difficult to |
287 |
find an employer willing to spend the money it takes to provide the |
288 |
employees with offices. |
289 |
|
290 |
When you consider the enormous amount of resources that are wasted for |
291 |
commuting in an economy and that some economies might start to gain an |
292 |
advantage over others by letting ppl work from their homes and by thus |
293 |
becoming able to make more competitive offers to their customers, you |
294 |
might come to think that it won't take very long before almost everyone |
295 |
must work at their home. So this isn't a problem of a company, or some |
296 |
companies, it's a problem for all companies and all employees, as it is |
297 |
a problem for all economies and all countries. |
298 |
|
299 |
>> >> It might work in theory. How would it be feasible in practise? |
300 |
>> > |
301 |
>> > Plenty of companies do it this way. If you don't want to pay for software |
302 |
>> > like XenDesktop, you need to do all the work setting it up yourself. |
303 |
>> |
304 |
>> VNC is somewhat slow over a 1Gbit LAN. Did they find some way to |
305 |
>> overcome this problem? |
306 |
> |
307 |
> Depends on the settings. |
308 |
|
309 |
Well, yes, I guess you can send something like 640x480 with some minimum |
310 |
content that changes as little as possible with less trouble over an |
311 |
internet connection than something one can actually work with. |
312 |
|
313 |
>> This sounds like it is for people with unlimited resources. |
314 |
>> |
315 |
>> BTW, access a VM through VNC, and you don't even have any way to make |
316 |
>> the mouse pointer in the VNC window actually follow the mouse pointer |
317 |
>> you're using, which makes it rather annoying to do anything in the VM |
318 |
>> you're looking at. If you found a solution for that, I'd be curious as |
319 |
>> to how you solved this problem. |
320 |
> |
321 |
> There is, it's even documented. |
322 |
> I'm assuming you are talking about the VNC-console Xen provides? |
323 |
> |
324 |
> Configure the mouse to be a tablet in the VM config and the issue disappears. |
325 |
|
326 |
Thanks, I can try that. I haven't seen this documented anywhere yet. |