<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div style="font-family: Calibri, Arial, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<span style="margin:0px;font-size:12pt">Thks, </span>
<div style="margin:0px;font-size:12pt"><br>
</div>
<div style="margin:0px;font-size:12pt">I'm new to Bird and certainly Debian so pls bear with me.</div>
<div style="margin:0px;font-size:12pt"><br>
</div>
<div style="margin:0px;font-size:12pt">So below shows a number of Debian network stack tweaks for your specific use case, is that fair ?</div>
<div style="margin:0px;font-size:12pt">With kernel tweaks to follow from <span style="margin:0px;font-size:15px;background-color:rgb(255, 255, 255);display:inline !important">Cybertinus</span></div>
<div style="margin:0px;font-size:12pt"><span style="margin:0px;font-size:15px;background-color:rgb(255, 255, 255);display:inline !important"><br>
</span></div>
<div style="margin:0px;font-size:12pt"><span style="margin:0px;font-size:15px;background-color:rgb(255, 255, 255);display:inline !important">@<span style="margin:0px;background-color:rgb(255, 255, 255);display:inline !important">Cybertinus - That would be appreciated. </span></span></div>
<div style="margin:0px;font-size:12pt"><br>
</div>
<div style="margin:0px;font-size:12pt"><br>
</div>
<div style="margin:0px;font-size:12pt">FYI</div>
<div style="margin:0px;font-size:12pt"> Our setup is Anycast based Bird stacks distributed globally linking directly from virtual image to DC hosting provider who we peer with using BGP, with own ASN & IPv4/24 & IPV6/28. Routing wise 1 IPV4 subnet & 1 IPV6
advertised out to DC Host provider to announce on Internet. So basic routing table. Our host address will grow over time up to max 253/4 ish running as dummy entries (in Centos) which are utilised by HAproxy. So a corner case that appears relatively simple
routing Bird wise. Appreciate any thoughts etc.</div>
<div style="margin:0px;font-size:12pt"><br>
</div>
<div style="margin:0px;font-size:12pt"><br>
</div>
<div style="margin:0px;font-size:12pt">Will</div>
<span style="margin:0px;font-size:12pt"></span><br>
</div>
<div>
<div id="appendonsend"></div>
<div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
<br>
</div>
<hr tabindex="-1" style="display:inline-block; width:98%">
<div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" color="#000000" style="font-size:11pt"><b>From:</b> Kees Meijs | Nefos <kees@nefos.nl><br>
<b>Sent:</b> 17 March 2021 10:59<br>
<b>To:</b> W Templeton <wnt@btconnect.com>; BIRD Users <bird-users@network.cz><br>
<b>Cc:</b> Will Templeton <wtempleton@tbscg.com><br>
<b>Subject:</b> Re: Bird Centos 8 Replacement recommendations request</font>
<div> </div>
</div>
<div>Hi,<br>
<br>
<div class="x_moz-cite-prefix">On 17-03-2021 11:53, W Templeton wrote:<br>
</div>
<blockquote type="cite">
<div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
1.your OS version <br>
</div>
</blockquote>
<br>
Sure:<br>
<blockquote type="cite">$ cat /etc/debian_version <br>
10.8<br>
</blockquote>
<br>
Basicly: Debian stable with current updates.<br>
<br>
<blockquote type="cite">
<div style="font-family:Calibri,Arial,Helvetica,sans-serif; font-size:12pt; color:rgb(0,0,0)">
2.Debian kernel tweaking</div>
</blockquote>
<br>
Sure:<br>
<blockquote type="cite"># cat 99-router.conf <br>
# Enable IP forwarding for IPv4 and IPv6 AF.<br>
net.ipv4.conf.all.forwarding = 1<br>
net.ipv6.conf.all.forwarding = 1<br>
<br>
# Send ICMP errors from receiving interface and be a nice netizen.<br>
net.ipv4.icmp_errors_use_inbound_ifaddr = 1<br>
<br>
# Disable reverse path filtering.<br>
net.ipv4.conf.all.rp_filter = 0<br>
net.ipv4.conf.default.rp_filter = 0<br>
<br>
# Allow more IPv6 routes.<br>
net.ipv6.route.max_size = 2147483647<br>
<br>
# Allow more network neighbours.<br>
net.ipv4.neigh.default.gc_thresh1=16384<br>
net.ipv4.neigh.default.gc_thresh2=32768<br>
net.ipv4.neigh.default.gc_thresh3=65536<br>
net.ipv6.neigh.default.gc_thresh1=16384<br>
net.ipv6.neigh.default.gc_thresh2=32768<br>
net.ipv6.neigh.default.gc_thresh3=65536<br>
</blockquote>
<br>
<blockquote type="cite"># cat 99-tengige.conf <br>
# 10GB/54MB (56623104)<br>
net.core.rmem_max = 56623104<br>
net.core.wmem_max = 56623104<br>
net.core.rmem_default = 56623104<br>
net.core.wmem_default = 56623104<br>
net.core.optmem_max = 40960<br>
net.ipv4.tcp_rmem = 4096 87380 56623104<br>
net.ipv4.tcp_wmem = 4096 65536 56623104<br>
<br>
# Increase number of incoming connections. The value can be raised to bursts of request, default is 128<br>
net.core.somaxconn = 1024<br>
<br>
# Increase number of incoming connections backlog, default is 1000<br>
net.core.netdev_max_backlog = 50000<br>
<br>
# Maximum number of remembered connection requests, default is 128<br>
net.ipv4.tcp_max_syn_backlog = 30000<br>
<br>
# Increase the tcp-time-wait buckets pool size to prevent simple DoS attacks, default is 8192<br>
net.ipv4.tcp_max_tw_buckets = 2000000<br>
<br>
# Recycle and Reuse TIME_WAIT sockets faster, default is 0 for both<br>
net.ipv4.tcp_tw_recycle = 1<br>
net.ipv4.tcp_tw_reuse = 1<br>
<br>
# Decrease TIME_WAIT seconds, default is 30 seconds<br>
net.ipv4.tcp_fin_timeout = 10<br>
<br>
# Tells the system whether it should start at the default window size only for TCP connections<br>
# that have been idle for too long, default is 1<br>
net.ipv4.tcp_slow_start_after_idle = 0<br>
<br>
# If your servers talk UDP, also up these limits, default is 4096<br>
net.ipv4.udp_rmem_min = 8192<br>
net.ipv4.udp_wmem_min = 8192<br>
</blockquote>
<br>
<blockquote type="cite"># cat 99-vrrp.conf <br>
# For VRRP.<br>
net.ipv4.ip_nonlocal_bind = 1<br>
</blockquote>
<br>
Cybertinus: maybe you can give insight in your kernel tweaks as well? (Probably very overlapping.)<br>
<br>
Cheers,<br>
Kees<br>
</div>
</div>
</body>
</html>