<html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=Windows-1252">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
.MsoChpDefault
{mso-style-type:export-only;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:72.0pt 72.0pt 72.0pt 72.0pt;}
div.WordSection1
{page:WordSection1;}
--></style>
</head>
<body lang="EN-IN" link="blue" vlink="#954F72" style="word-wrap:break-word">
<div class="WordSection1">
<p class="MsoNormal"><span lang="EN-US">Hi Paul,<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US"><o:p> </o:p></span></p>
<p class="MsoNormal"><span lang="EN-US">Thanks for your response. I understand the concern for good random numbers; but in this scenario when second thread calls SSL_CTX_new it is waiting forever in RAND_priv_bytes(). Looks like entropy functions defined by
first (bind) thread are very specific for its own use case and can’t be used by other treads.<o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US">So I am thinking of using default</span> OpenSSL RAND_METHOD for second thread and keep first thread (bind) to use its own random number generators.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Please let me know how can I make one thread use default RAND_METHOD and keep other thread to use its own method. I have gone through RAND_bytes() and drbg_bytes() but not getting enough idea. It would be really helpful if you point out
APIs which help me to achieve this requirement.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Thank You,<o:p></o:p></p>
<p class="MsoNormal">Vishwanath M<span lang="EN-US"><o:p></o:p></span></p>
<p class="MsoNormal"><span lang="EN-US"><o:p> </o:p></span></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<div style="mso-element:para-border-div;border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal" style="border:none;padding:0cm"><b>From: </b><a href="mailto:openssl-users-request@openssl.org">openssl-users-request@openssl.org</a><br>
<b>Sent: </b>03 April 2021 02:19 PM<br>
<b>To: </b><a href="mailto:openssl-users@openssl.org">openssl-users@openssl.org</a><br>
<b>Subject: </b>openssl-users Digest, Vol 77, Issue 6</p>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Send openssl-users mailing list submissions to<br>
openssl-users@openssl.org<br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="https://mta.openssl.org/mailman/listinfo/openssl-users">https://mta.openssl.org/mailman/listinfo/openssl-users</a><br>
or, via email, send a message with subject or body 'help' to<br>
openssl-users-request@openssl.org<br>
<br>
You can reach the person managing the list at<br>
openssl-users-owner@openssl.org<br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of openssl-users digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. Re: openssl-users Digest, Vol 77, Issue 4 (Dr Paul Dale)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Sat, 3 Apr 2021 18:48:48 +1000<br>
From: Dr Paul Dale <pauli@openssl.org><br>
To: openssl-users@openssl.org<br>
Subject: Re: openssl-users Digest, Vol 77, Issue 4<br>
Message-ID: <c1d66d09-56da-8d62-96ec-2286b1b65534@openssl.org><br>
Content-Type: text/plain; charset="windows-1252"; Format="flowed"<br>
<br>
I would be **very** concerned about bypassing a blocking RAND.? It is <br>
almost certainly blocking because it does not have enough randomness to <br>
satisfy your request.? By skipping this, you are likely getting poor <br>
quality random values and this can effectively negate any security you <br>
are gaining from the encryption.<br>
<br>
Good random numbers are fundamental to modern cryptography.? Without <br>
them, there is no security.? I cannot stress this enough.? Do not try to <br>
second guess or bypass the RNG.<br>
<br>
<br>
Pauli<br>
<br>
On 3/4/21 6:41 pm, Vishwanath Mahajanshetty wrote:<br>
><br>
> Thank You Paul and Matthias for your help.<br>
><br>
> The reason I am trying to have separate RAND_METHOD for two threads <br>
> is, the first thread which runs DNS *bind* code registers for <br>
> RAND_METHOD through dnssec module in it. It registers via either <br>
> ENGINE_set_default_RAND() or RAND_set_rand_method() based on <br>
> OPENSSL_NO_ENGINE is defined or not. But problem is, under some <br>
> circumstances the random number generator enters into blocking mode <br>
> and starts to wait for some events on some FDs and it blocks in <br>
> select() system call. dst__entropy_getdata() ?from bind code is doing <br>
> this. I am not sure under what cases it enters into blocking mode.<br>
><br>
> So If I use this RND_METHOD in second thread (basically this thread <br>
> does different task of handling *DoT*, Dns Over TLS, connections, <br>
> which is not related to first thread wrt SSL functionalities), then <br>
> while creating SSL_CTX this thread gets stuck in select() system call <br>
> randomly (happens very rarely as decided by dst__entropy_getdata()); <br>
> this can happen at any time of SSL connection lifetime whenever it <br>
> wants to get random data.<br>
><br>
> I agree with you that we should have done this as separate process <br>
> instead of new thread; but I am trying figure out if I can somehow <br>
> avoid this situation.<br>
><br>
> As you mentioned, I tried to look into implementation of RAND_bytes() <br>
> and drbg_bytes().<br>
><br>
> When SSL_CTX_new() calls RAND_bytes(), it calls RAND_get_rand_method() <br>
> which returns RAND_METHOD set by *bind* thread. So if I avoid <br>
> configuring RAND_METHOD in *bind* thread, then RAND_get_rand_method() <br>
> will return *rand_meth *which is OpenSSL default RAND_METHOD; but if I <br>
> do this change bind thread will move away from its RAND_METHOD <br>
> functions and start using OpenSSL default functions which may change <br>
> its behaviour.<br>
><br>
> So I am still confused how can I do *bind* thread to use its own <br>
> RAND_METHOD and *DoT* thread to use default OpenSSL RAND_METHOD. It <br>
> would be really helpful if you can explain this with little more <br>
> details (are there any APIs I can call from one thread to use its <br>
> specific RAND_METHOD but other threads continue to use OpenSSL default <br>
> RAND_METHOD?).<br>
><br>
> Thank You,<br>
><br>
> Vishwanath M<br>
><br>
> *From: *openssl-users-request@openssl.org <br>
> <<a href="mailto:openssl-users-request@openssl.org">mailto:openssl-users-request@openssl.org</a>><br>
> *Sent: *02 April 2021 04:58 PM<br>
> *To: *openssl-users@openssl.org <<a href="mailto:openssl-users@openssl.org">mailto:openssl-users@openssl.org</a>><br>
> *Subject: *openssl-users Digest, Vol 77, Issue 4<br>
><br>
> Send openssl-users mailing list submissions to<br>
> ??????? openssl-users@openssl.org<br>
><br>
> To subscribe or unsubscribe via the World Wide Web, visit<br>
> <a href="https://mta.openssl.org/mailman/listinfo/openssl-users">https://mta.openssl.org/mailman/listinfo/openssl-users</a>
<br>
> <<a href="https://mta.openssl.org/mailman/listinfo/openssl-users">https://mta.openssl.org/mailman/listinfo/openssl-users</a>><br>
> or, via email, send a message with subject or body 'help' to<br>
> ??????? openssl-users-request@openssl.org<br>
><br>
> You can reach the person managing the list at<br>
> ??????? openssl-users-owner@openssl.org<br>
><br>
> When replying, please edit your Subject line so it is more specific<br>
> than "Re: Contents of openssl-users digest..."<br>
><br>
><br>
> Today's Topics:<br>
><br>
> ?? 1. Re: Regarding RAND_set_rand_method (Dr Paul Dale)<br>
> ?? 2. RE: Regarding RAND_set_rand_method (Dr. Matthias St. Pierre)<br>
><br>
><br>
> ----------------------------------------------------------------------<br>
><br>
> Message: 1<br>
> Date: Fri, 2 Apr 2021 16:51:28 +1000<br>
> From: Dr Paul Dale <pauli@openssl.org><br>
> To: openssl-users@openssl.org<br>
> Subject: Re: Regarding RAND_set_rand_method<br>
> Message-ID: <1781ab4c-2e2b-fa3b-8b3c-fb4fc5bd3371@openssl.org><br>
> Content-Type: text/plain; charset="windows-1252"; Format="flowed"<br>
><br>
> There isn't an easy a way to do what you want in 1.1.1.<br>
> RAND_set_rand_method replaces the RNG for all of OpenSSL.? In theory<br>
> your RAND_METHOD could detect which thread it is running in and do<br>
> different things for each.? I'm not sure this is a good idea however.<br>
><br>
> Why aren't the random number from your first thread good enough for the<br>
> second?? Good random numbers are just that - random.? It should be<br>
> impossible to distinguish the two streams.<br>
><br>
> In OpenSSL 3.0 there are ways to achieve what you're wanting.<br>
><br>
><br>
> Pauli<br>
><br>
> On 2/4/21 4:24 pm, Vishwanath Mahajanshetty wrote:<br>
> ><br>
> > Hi,<br>
> ><br>
> > I have some doubts/questions on how to use methods (for ex:<br>
> > RAND_set_rand_method) in multi threaded application which use OpenSSL.<br>
> > In my application (running on OpenSSL 1.1.1d) there are two threads<br>
> > which use OpenSSL, both threads perform very different operations. The<br>
> > issue I am facing is as below:<br>
> ><br>
> > Thread T1 calls RAND_set_rand_method() and sets RAND_METHOD structure.<br>
> > This is very specific to T1s use case. When thread T2 wants to create<br>
> > SSL_CTX it calls SSL_CTX_new() which then calls RAND_priv_bytes(). I<br>
> > am observing that the function RAND_priv_bytes() is calling the<br>
> > function set by T1 by RAND_METHOD in RAND_set_rand_method().<br>
> ><br>
> > Essentially RAND_METHOD function set by thread T1 are getting called<br>
> > by thread T2.<br>
> ><br>
> > *Q1: I want to know is there any way to avoid this problem? I want<br>
> > thread T2 to call default RAND methods and avoid calling methods set<br>
> > by thread T1. This is not only for RAND methods, but for any other<br>
> > methods.*<br>
> ><br>
> > **<br>
> ><br>
> > Q2: Also, is it possible to run OpenSSL as separate instance per<br>
> > thread (where each thread can do its own OpenSSL initialization) so<br>
> > that they can avoid above mentioned problem?<br>
> ><br>
> > Thank you,<br>
> ><br>
> > Vishwanath M<br>
> ><br>
><br>
> -------------- next part --------------<br>
> An HTML attachment was scrubbed...<br>
> URL: <br>
> <https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/58bcb71b/attachment-0001.html
<br>
> <<a href="https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/58bcb71b/attachment-0001.html">https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/58bcb71b/attachment-0001.html</a>>><br>
><br>
> ------------------------------<br>
><br>
> Message: 2<br>
> Date: Fri, 2 Apr 2021 11:27:53 +0000<br>
> From: "Dr. Matthias St. Pierre" <Matthias.St.Pierre@ncp-e.com><br>
> To: Dr Paul Dale <pauli@openssl.org>, "openssl-users@openssl.org"<br>
> ??????? <openssl-users@openssl.org><br>
> Subject: RE: Regarding RAND_set_rand_method<br>
> Message-ID: <7056523443ae4f94bca32240c4f24533@ncp-e.com><br>
> Content-Type: text/plain; charset="us-ascii"<br>
><br>
> Re Q1: I want to know is there any way to avoid this problem? I want <br>
> thread T2 to call default RAND methods and avoid calling methods set <br>
> by thread T1. This is not only for RAND methods, but for any other <br>
> methods.<br>
><br>
> First of all, I agree with Pauli: your first question should be, why <br>
> do you need different random generators for different threads in the <br>
> same application? Is this necessary, or are you overengineering?<br>
><br>
> Let me clarify some details about the RNG implemention in OpenSSL <br>
> 1.1.1.: The RAND_METHOD interface itself is not thread aware. It is <br>
> only the new default RAND_METHOD implementation (added in 1.1.1.) of <br>
> OpenSSL (RAND_OpenSSL()), which supports thread local random <br>
> generators. The implementation is based on deterministic random bit <br>
> generators (DRBG) as described in NIST.SP.800-90Ar1. Wenn a thread <br>
> calls RAND_bytes() (resp. RAND_priv_bytes()), the call is forwarded to <br>
> the thread-specific DRBG instance. All per-thread instances reseed <br>
> from a single global DRBG instance, which in turn reseeds from? from <br>
> random sources provided by the operating system.<br>
><br>
> In your case, by replacing the RAND_METHOD, you are changing the <br>
> complete RAND implementation for all threads. Moreover, you are <br>
> completely responsible yourself for reseeding your RNG properly.<br>
><br>
> You could however implement a smarter RAND_METHOD which calls your <br>
> specific RNG for T1 and delegates to the thread local DRBG <br>
> (RAND_DRBG_get0_public() resp. RAND_DRBG_get0_private()) for all other <br>
> threads. To get an idea how it can be done, take a look at the default <br>
> implementation of RAND_bytes(), drbg_bytes() in drbg_lib.c:<br>
><br>
> <a href="https://github.com/openssl/openssl/blob/OpenSSL_1_1_1-stable/crypto/rand/drbg_lib.c#L958-L970">
https://github.com/openssl/openssl/blob/OpenSSL_1_1_1-stable/crypto/rand/drbg_lib.c#L958-L970</a>
<br>
> <<a href="https://github.com/openssl/openssl/blob/OpenSSL_1_1_1-stable/crypto/rand/drbg_lib.c#L958-L970">https://github.com/openssl/openssl/blob/OpenSSL_1_1_1-stable/crypto/rand/drbg_lib.c#L958-L970</a>><br>
><br>
><br>
> Re Q2: Also, is it possible to run OpenSSL as separate instance per <br>
> thread (where each thread can do its own OpenSSL initialization) so <br>
> that they can avoid above mentioned problem?<br>
><br>
> No. If you really need something like that, you might want to consider <br>
> splitting your two threads into two processes.<br>
><br>
> HTH,<br>
> Matthias<br>
><br>
><br>
><br>
> From: openssl-users <openssl-users-bounces@openssl.org> On Behalf Of <br>
> Dr Paul Dale<br>
> Sent: Friday, April 2, 2021 8:51 AM<br>
> To: openssl-users@openssl.org<br>
> Subject: Re: Regarding RAND_set_rand_method<br>
><br>
> There isn't an easy a way to do what you want in 1.1.1. <br>
> RAND_set_rand_method replaces the RNG for all of OpenSSL.? In theory <br>
> your RAND_METHOD could detect which thread it is running in and do <br>
> different things for each.? I'm not sure this is a good idea however.<br>
><br>
> Why aren't the random number from your first thread good enough for <br>
> the second?? Good random numbers are just that - random.? It should be <br>
> impossible to distinguish the two streams.<br>
><br>
> In OpenSSL 3.0 there are ways to achieve what you're wanting.<br>
><br>
><br>
> Pauli<br>
> On 2/4/21 4:24 pm, Vishwanath Mahajanshetty wrote:<br>
> Hi,<br>
><br>
> I have some doubts/questions on how to use methods (for ex: <br>
> RAND_set_rand_method) in multi threaded application which use OpenSSL. <br>
> In my application (running on OpenSSL 1.1.1d) there are two threads <br>
> which use OpenSSL, both threads perform very different operations. The <br>
> issue I am facing is as below:<br>
><br>
> Thread T1 calls RAND_set_rand_method() and sets RAND_METHOD structure. <br>
> This is very specific to T1s use case. When thread T2 wants to create <br>
> SSL_CTX it calls SSL_CTX_new() which then calls RAND_priv_bytes(). I <br>
> am observing that the function RAND_priv_bytes() is calling the <br>
> function set by T1 by RAND_METHOD in RAND_set_rand_method().<br>
><br>
> Essentially RAND_METHOD function set by thread T1 are getting called <br>
> by thread T2.<br>
><br>
> Q1: I want to know is there any way to avoid this problem? I want <br>
> thread T2 to call default RAND methods and avoid calling methods set <br>
> by thread T1. This is not only for RAND methods, but for any other <br>
> methods.<br>
><br>
> Q2: Also, is it possible to run OpenSSL as separate instance per <br>
> thread (where each thread can do its own OpenSSL initialization) so <br>
> that they can avoid above mentioned problem?<br>
><br>
> Thank you,<br>
> Vishwanath M<br>
><br>
><br>
> -------------- next part --------------<br>
> An HTML attachment was scrubbed...<br>
> URL: <br>
> <https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/53153b3a/attachment.html
<br>
> <<a href="https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/53153b3a/attachment.html">https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/53153b3a/attachment.html</a>>><br>
> -------------- next part --------------<br>
> A non-text attachment was scrubbed...<br>
> Name: smime.p7s<br>
> Type: application/pkcs7-signature<br>
> Size: 7494 bytes<br>
> Desc: not available<br>
> URL: <br>
> <https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/53153b3a/attachment.bin
<br>
> <<a href="https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/53153b3a/attachment.bin">https://mta.openssl.org/pipermail/openssl-users/attachments/20210402/53153b3a/attachment.bin</a>>><br>
><br>
> ------------------------------<br>
><br>
> Subject: Digest Footer<br>
><br>
> _______________________________________________<br>
> openssl-users mailing list<br>
> openssl-users@openssl.org<br>
> <a href="https://mta.openssl.org/mailman/listinfo/openssl-users">https://mta.openssl.org/mailman/listinfo/openssl-users</a>
<br>
> <<a href="https://mta.openssl.org/mailman/listinfo/openssl-users">https://mta.openssl.org/mailman/listinfo/openssl-users</a>><br>
><br>
><br>
> ------------------------------<br>
><br>
> End of openssl-users Digest, Vol 77, Issue 4<br>
> ********************************************<br>
><br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <<a href="https://mta.openssl.org/pipermail/openssl-users/attachments/20210403/e6d569b8/attachment.html">https://mta.openssl.org/pipermail/openssl-users/attachments/20210403/e6d569b8/attachment.html</a>><br>
<br>
------------------------------<br>
<br>
Subject: Digest Footer<br>
<br>
_______________________________________________<br>
openssl-users mailing list<br>
openssl-users@openssl.org<br>
<a href="https://mta.openssl.org/mailman/listinfo/openssl-users">https://mta.openssl.org/mailman/listinfo/openssl-users</a><br>
<br>
<br>
------------------------------<br>
<br>
End of openssl-users Digest, Vol 77, Issue 6<br>
********************************************<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
</body>
</html>